ISSN 1195-8898 .

CMOS

Canadian Meteorological and Oceanographic Society BULLETIN SCMO La Société canadienne de météorologie et June / juin 2008 Vol.36 No.3 d'océanographie

DRI Drought Research Initiative Initiative de recherche sur la sécheresse CMOS Bulletin SCMO "at the service of its members / au service de ses membres" Canadian Meteorological and Oceanographic Society (CMOS)

Editor / Rédacteur: Paul-André Bolduc Associate Editor / Rédactrice associée: Dorothy Neale Société canadienne de météorologie Canadian Meteorological and Oceanographic Society et d'océanographie (SCMO) Société canadienne de météorologie et d’océanographie P.O. Box 3211, Station D Ottawa, ON, Canada K1P 6H7 E-Mail: [email protected]; Courriel: [email protected] Executive / Exécutif (until 30 May 2008) President / Président Dr. Paul G. Myers University of Alberta This issue : Large-area, prolonged droughts are among Tel: 780-492-6706; Fax: 780-492-2030 Canada’s costliest natural disasters having major impacts E-mail/Courriel: [email protected] on a wide range of sectors including agriculture, forestry, industry, municipalities, recreation, health and society, and Vice-President / Vice-président aquatic ecosystems. Although most regions of Canada Andrew Bush experience drought, the Canadian Prairies are most University of Alberta susceptible with at least 40 droughts being documented Tel: 780-492-0351; Fax: 780-492-2030 during the past two centuries. However, rarely has drought E-mail/Courriel: [email protected] been as serious or extensive as the recent 1999-2005 episode that produced the worst drought for at least a Treasurer / Trésorier hundred years in many parts of the Canadian Prairies. To Ron Hopkinson aid in addressing such extreme drought events, the Drought Custom Climate Services, Regina Research Initiative (DRI) research network has been Tel: 306-586-5489; Fax: 306-586-5489 established. The focus of DRI is to better understand the E-mail/Courriel: [email protected] factors that led to, sustained and ended this recent drought with a view to understanding its structure and to contributing Corresponding Secretary / Secrétaire-correspondant to the better anticipation of such events. To accomplish Bob Kochtubajda this objective, the drought is being considered from several Environment Canada perspectives involving atmospheric, surface (including Tel: 780-951-8811; Fax: 780-951-8634 vegetation) and sub-surface processes. Furthermore the E-mail/Courriel: [email protected] DRI research community is working closely with many partners affected by the drought so that they can better Recording Secretary / Secrétaire d'assemblée cope with such events in the future. Bill Hume Tel: 780-989-4103 Three articles in this issue address Canadian droughts E-mail/Courriel: [email protected] with a focus on the ongoing DRI Network. An overview of droughts in Canada is initially presented on page 79 with an Past-President / Président ex-officio emphasis on those occurring over the Canadian Prairies. Dr. Geoff Strong This is followed by a description on page 87 of the many Tel: 780-922-0665; Fax: 780-922-0678 facets of DRI. The issue concludes with an assessment of E-mail/Courriel: [email protected] the user community’s expectation of DRI including their identified needs on page 97. Councillors-at-large / Conseillers 1) Dr. Neil Campbell Cover page: The cover page photo was taken near Consul, Tel: 613-731-4512 Saskatchewan in April 2002. Photo courtesy of Ted Banks, E-mail/Courriel: [email protected] Agriculture and Agri-Food Canada. 2) Kent Johnson Environment Canada, Kelowna Note du Rédacteur: Vous trouverez la description Tel: 604-763-3532 française de ce numéro thématique ainsi que de la E-mail/Courriel: [email protected] page couverture à la page 102. 3) Brad Shannon Environment Canada, Calgary Tel: 403-299-3534; E-mail/Courriel: [email protected]

-ii- ....from the Presidents’ Desks Volume 36 No.3 June 2008 Juin 2008 Friends and colleagues: — Inside / En Bref Now that June has arrived, I am amazed at from the Presidents’ desks how fast the past year Allocution des présidents has gone by and then my by/par Paul Myers & Andy Bush page 77 time as CMOS President has come to an end. In Highlights of Recent CMOS Meetings page 78 some ways this is a relief in that I can go back to Articles focussing on some Droughts in Canada: An Overview interesting research by Barrie Bonsal page 79 questions that I'd like to look at. Yet, I am also A Drought Research Initiative for the disappointed that this Canadian Prairies by Ronald Stewart, year is over so fast John Pomeroy and Rick Lawford page 87 because of how much I have learned about User Expectations for the Drought Research aspects of science, Initiative (DRI) by Rick Lawford, Harvey Hill, especially on the Elaine Wheaton, Irene Hanuta, governmental and Alf Warkentin and Bill Girling page 97 Paul Myers management side, and Outgoing President of CMOS how much more I'd like to Improved seawater thermodynamics: How should the Président sortant de la SCMO learn and be involved in proposed change in salinity be implemented by different things that I only SCOR/IAPSO Working Group 127 page 103 got a chance to touch on during the past year. Our regular sections / Nos chroniques régulières

And I do think this last issue is important. As CMOS Business / Affaires de la SCMO page 107 scientists/forecasters, most of us want to spend their time focussing on what we all love, our favourite aspects of In Memoriam page 110 meteorology and oceanography. Yet, addressing the issues that impact our field, as well as communicating our science Short News / Nouvelles brèves page 111 to the general public are issues of great importance. And it is something that I suspect almost all of us can do better - CMOS Accredited Consultants / I sure learnt that during this year as CMOS President. And Experts-conseils accrédités de la SCMO page 112 one way for this to occur is to become more involved with Printed in Kanata, Ontario, by Gilmore Printing Services Inc. CMOS, as the society continues to represent and promote Imprimé sous les presses de Gilmore Printing Services Inc., Kanata, meteorology and oceanography in Canada. And thus, Ontario. echoing all previous CMOS presidents in their final words, I highly recommend members to get involved as CMOS volunteers in your Centres, and on national committees or This publication is produced under the authority of the Canadian Council. Meteorological and Oceanographic Society. Except where explicitly stated, opinions expressed in this publication are those of the authors and are not necessarily endorsed by the Society. Finally, I would like to thank those in the national office in Ottawa, especially our executive director Ian Rutherford, Cette publication est produite sous la responsabilité de la Société without whom I would not have survived this year, and canadienne de météorologie et d’océanographie. À moins d’avis without which the society would not function. And so to contraire, les opinions exprimées sont celles des auteurs et ne conclude I wish the best for our new incoming president reflètent pas nécessairement celles de la Société. Andy Bush as I prepare to pass off the position to him.

Paul Myers

Outgoing President / Président sortant Canadian Meteorological and Oceanographic Society Société canadienne de météorologie et d’océanographie CMOS exists for the advancement of meteorology and oceanography in Canada.

Le but de la SCMO est de stimuler l'intérêt pour la (Continued on next page / Suite à la page suivante) météorologie et l'océanographie au Canada.

-77- CMOS Bulletin SCMO Vol.36, No.3 ...from the Presidents’ desks (Continued & suite) Highlights of Recent CMOS Meetings

March Council Meeting

# Updating status of Congresses: Kelowna, 2008, Halifax, 2009 and Ottawa, 2010; # Discussing the relevent agreements needed between CMOS and CGU for the 2010 congress to be joint between the two societies; # Updates on new membership applications and society membership numbers; # Overview of the status of the CMOS Tour Speaker this year - what had occurred and which talks were still to be given; # Council approval of nominees for awards from the Andy Bush Prizes and Awards Committee; Incoming President of CMOS # Council approval of Fellow candidates from the Fellows Nouveau président de la SCMO Committee; # Discussion of issues related to awards, membership Being literally next door to Paul Myers, I can attest to the and getting sufficient nominations for some awards (i.e. amount of effort he has put in as President of CMOS. I think not all CMOS awards will be awarded this year we owe Paul a huge collective thanks for his efforts over the because of a lack of nominations, even with extensions past year and, Paul, have one fantastic and well-earned of the deadlines); # sabbatical! I would also like to thank and pay my respects Appointment of CMOS Audit Committee; # to Ian Rutherford, for steering things in the right way, at the Discussion of the possibility of having flight service right time. specialists become part of CMOS, with CMOS governing their accreditation process; # The next few years will be challenging indeed. With the Approval of an update policy on travel expenses; # potential end of funding through the Canadian Foundation Presentation of a discussion paper on congress for Climate and Atmospheric Sciences (CFCAS) it is not finances, with a decision to use the Halifax congress clear what avenue researchers in Atmospheric, Oceanic, or (with the agreement of the Halifax LAC) as a test bed Cryospheric Sciences might pursue given NSERC's woeful for dealing with more of the finances at a national level; # history for Environmental Sciences reallocations. One Presentation of a discussion paper related to CMOS' hopeful pursuit will be continuing our link with the Canadian financial investments and what rate of return the Geophysical Union (CGU) and I am optimistic that we can society should look to get, considering the needs to form a unified front for geophysics, atmospheric physics, preserve the capital versus growth; # oceanography and cryospheric dynamics. Our joint meeting Discussion of the society draft budget for 2008-09 prior last year in St. John's, NF, was in my opinion an enormous to the AGM at Kelowna; success and I wish to pursue this link with the current Paul Myers, president of CGU, John Pomeroy. CMOS President Président de la SCMO The collective public appears to be concerned about climate change and impacts on Canada. We are the natural organization to speak to this issue, to educate the public, to educate the political parties, and to educate industry. The latter two are difficult, of course. With Arctic climate becoming such an ironically hot topic we need to be at the URGENT - URGENT - URGENT - URGENT forefront of educating those people who will be relevant players in the future of our North. This is not an easy task CMOS Bulletin SCMO but I'll try, in the coming year, to do my best. Next Issue

Andy Bush Next issue of the CMOS Bulletin SCMO will be published in August 2008. Please send your articles, notes, Incoming President / Nouveau président workshop reports or news items before July 4, 2008 to Canadian Meteorological and Oceanographic Society the address given on page ii. We have an URGENT need Société canadienne de météorologie et d’océanographie for your written contributions.

CMOS Bulletin SCMO Vol.36, No.3 -78- ARTICLES

Droughts in Canada: An Overview

by Barrie Bonsal1

Résumé: Les sécheresses menacent sérieusement la société et l’environnement puisqu’en grande partie les activités humaines et la santé des écosystèmes sont dépendantes de l’apport suffisant en eau. Les sécheresses prolongées sur grande surface sont parmi les catastrophes naturelles les plus coûteuses au Canada. Elles ont des répercussions profondes sur tout un éventail de secteurs, notamment sur l’agriculture, les forêts, l’industrie, les municipalités, les loisirs, la santé, la société et les écosystèmes aquatiques. Bien que la plupart des régions du Canada doivent faire face à la sécheresse, ce sont les régions du sud des Prairies canadiennes qui sont les plus vulnérables en raison de leurs précipitations spatio-temporelles éminemment variables. Cet article donne une vue d’ensemble des sécheresses au Canada et met l’accent sur les Prairies canadiennes. Tout d’abord, avec l’appui des données des instruments et des registres anciens, on fait une revue des tendances du passé et de la variabilité de la présence des sécheresses à travers le pays. Par la suite, on décrit les connaissances existantes en rapport avec les causes atmosphériques à grande échelle reliées à la sécheresse canadienne. On y discute aussi des sécheresses futures. On y présente les techniques de contrôles actuelles, les capacités de la modélisation et de la prévision, et les stratégies d’adaptation des sécheresses canadiennes. En conclusion, l’article met l’accent sur l’identification des lacunes majeures de la recherche et des besoins de programmes concernant les sécheresses sur l’Amérique du Nord qui pourraient nous aider à comprendre et prévoir leur présence, à contrôler et modéliser leur état, et à s’adapter à leurs effets négatifs.

Introduction Droughts are one of the most dramatic manifestations of since drought is characteristic of dry environments variations in the water cycle. Prolonged, large-area (Maybank et al., 1995). Droughts occur on a variety of droughts are among Canada’s costliest natural disasters temporal and spatial scales with their impacts being having major impacts on a wide range of sectors including dependent on the timing and sequencing of dry periods. For agriculture, forestry, industry, municipalities, recreation, example, a shortage of water and soil moisture at a critical human health and society, and aquatic ecosystems. They time for crop growth may initiate agricultural drought, but frequently stress water availability by depleting soil hydropower generation would not be affected if reservoirs moisture, reducing stream flows, lowering lake and reservoir have adequate supplies. Climate anomalies that last from levels, and diminishing groundwater supplies. This a month to years are the root of most droughts, however, ultimately affects several economic activities including, for human impacts on resources and climate and the changing example, decreased agricultural production, less hydro- demand for water are also major contributing factors electric power generation and increased freshwater (McKay et al., 1989). transportation costs. Droughts also create major environmental hazards such as reduced water quality, Although most regions of Canada have experienced wetland loss, soil erosion and degradation and ecological drought, southern regions of the Canadian Prairies and habitat destruction. Moreover, in response to the economic interior British Columbia are more susceptible mainly due to and environmental significance of droughts, scientific their high variability of precipitation in time and space. concern has been expressed regarding climate-change During the past two centuries, at least 40 droughts have impacts on future drought frequency, duration and severity occurred in western Canada with multi-year episodes being over various regions of the globe including Canada. observed in the 1890s, 1910s, 1930s, 1960s, and 1980s (Phillips, 1990; Chipanshi et al., 2006). Droughts in Droughts are complex phenomena with no standard southern Ontario/Quebec are usually shorter, smaller in definition. Simply stated, drought is a prolonged period of area, less frequent and less intense. Nonetheless, there abnormally dry weather that depletes water resources for have been some major drought occurrences during the 20th human and environmental needs (MSC Drought Study century. Droughts in the Atlantic Provinces occur even less Group, 1986). However, each drought is different depending frequently. This reduced occurrence results in lower on factors such as area affected, duration, intensity, adaptive capacity, thus making the region more susceptible antecedent conditions and the region’s capability to adapt to drought impacts. Droughts are less of a concern for to water shortages. Droughts also differ from other disasters northern Canada mainly due to lower population densities. (e.g., floods) since they have long durations, and lack easily However, increased frequencies of forest fires during identified onsets and terminations. Furthermore, their drought years can have serious economic impacts. recurrence in drought-prone areas is practically certain

1 Environment Canada, Saskatoon, Saskatchewan, Canada

-79- CMOS Bulletin SCMO Vol.36, No.3 Drought has, and continues to be a major concern in difficult. Examples of trends and variability in various Canada but rarely has it been as serious or extensive as the drought-related parameters are provided below. recent 1999-2004 episode. This event produced the worst drought for at least a hundred years in parts of Canada and High surface temperatures can intensify drought conditions in particular, the Canadian Prairies. Well below normal through enhanced evapotranspiration in summer, and precipitation was reported in areas of Alberta and increased sublimation and melting of the snowpack during Saskatchewan for more than four consecutive years winter. Several studies have identified significant trends in extending from autumn 1999 to spring 2004. Even in the temperature and various temperature-related indices over dust bowl of the 1930s, no single year between Medicine Canada during the 20th century. Mean annual air Hat, Kindersley, and Saskatoon was drier than in 2001. temperature has increased by an average of 0.9°C over Astonishingly, Saskatoon was 30% drier this year than any southern Canada for the period 1900-98. The greatest other over the last 110 years (Phillips, 2002). This recent warming was observed in the west and the largest rates drought was also unusual in terms of its vast spatial extent. occurred during winter and spring (Zhang et al., 2000). In particular, the years 2001 and 2002 may have brought Much of the country has also experienced significant trends the first coast-to-coast droughts on record, and were rare as toward longer frost-free periods (Bonsal et al., 2001b) which they struck areas that are less accustomed to dealing with could impact drought occurrence since these trends droughts including parts of Atlantic Canada and the translate into a longer ice-free season for lakes and rivers northern agricultural prairies. thus increasing the potential for open-water evaporation. From 1900-98, annual precipitation has significantly Unlike previous droughts that impacted few sectors, the increased over most of southern Canada with the exception 1999-2005 event affected many sectors including of southern Alberta and Saskatchewan. This pattern is agriculture, recreation, tourism, health, hydro-electricity and generally evident during all seasons (Zhang et al., 2000). forestry. Long-lasting impacts included soil degradation by The period 1915-97 was associated with substantial inter- wind erosion and deterioration of grasslands that could take decadal variability in North American snow cover including decades and longer to recover. Canada’s Gross Domestic lowest anomalies in the 1920s and 1930s and highest Product fell $5.8 billion for 2001 and 2002, with the larger during the late 1970s and early 1980s. Coincident with the loss in 2002 at more than $3.6 billion. Drought also large increases in spring temperature, the 1980s/90s were contributed to a negative net farm income in several characterized by rapid reductions in snow cover during the provinces for the first time in 25 years (Wheaton et al., second half of the snow season and especially during April 2005). Previously reliable water supplies such as streams, (Brown, 2000). wetlands, reservoirs, and groundwater were placed under stress and often failed. For example, the number of natural Examples of 20th century Palmer Drought Severity Index Prairie ponds in May 2002 was the lowest on record while (PDSI) time series for various regions of the country are in 2001, Great Lakes-St. Lawrence water levels plunged to provided in Figure 1 (Skinner, 2002). The series show their lowest point in more than 30 years thereby significantly considerable decadal-scale variability with no long-term increasing marine transportation costs. In British Columbia trends discernible in any portion of the country. Most and Manitoba, hydro-electric generation was curtailed, southern regions of Canada, however, were associated with necessitating additional purchases of power from anomalously negative PDSI values (i.e., drought conditions) neighbouring jurisdictions. In 2002, the incidence of forest during the late 1990s to early 2000s. In a study using the fires in Alberta increased to five times the ten-year average Standardized Precipitation Index (SPI) and PDSI as drought while in summer 2003, populated regions of interior British indicators, Bonsal and Regier (2007) showed that the worst Columbia were stricken by drought-enhanced fires. and most prolonged Canadian Prairie-wide droughts during the period 1915 to 2002 occurred in the early part of the 20th Past Trends and Variability century (1920s and 1930s). Sauchyn and Skinner (2001) Past drought research in Canada has been fragmented and reconstructed July PDSI for the southwestern Canadian often based on severe drought occurrence (e.g., the Plains using tree ring chronologies dating back to 1597. Drought Research Initiative described later in this issue). Results showed that the 20th century lacked the prolonged Nonetheless, there has been some effort to define large- droughts of the 18th and 19th centuries when the PDSI was scale trends and variability in various drought-related consistently below zero for decades at a time. Clusters of indices during the period of instrumental record. A drought years in the series suggested the existence of a 20 problematic issue for most of these trend analyses is that to 25 year periodicity over this region. they have been carried out independently with limited attempts in deriving comprehensive results for the entire country. They also often differ in terms of starting dates for trend calculation, and with respect to initial conditions for determining drought indices. Furthermore, the limitation of the instrumental record to approximately the last 100 years, combined with sparse high-resolution paleo-climatic information in areas that are most prone to drought, makes inference of long-term trends in Canadian droughts very

CMOS Bulletin SCMO Vol.36, No.3 -80- Figure 1: Annual PDSI values for a) Kamloops, BC, b) Saskatoon, SK, c) Sherbrooke, QC, and d) Yarmouth, NS. Solid lines represent 10-year running means. (Source: Climate Research Branch, Meteorological Service of Canada, Environment Canada, Downsview, ON.)

There have been some analyses of trends and variability in amplitude ridge centred over the area. The ridging creates various water-related drought indicators over Canada, but ‘blocking conditions’ that displace cyclonic tracks away from these records tend to be much shorter. Over the last 30 to the area (e.g., Chakravarti, 1976). Drought conditions can 50 years, mean stream flow has decreased in many parts of also be initiated and/or accentuated during the cold season Canada with significant reductions in the south (Zhang et when a lack of precipitation results in lower than normal al., 2001). Great Lakes’ water levels have shown spring runoff and thus reduced stream flow and reservoir considerable decadal-scale variability during the 20th and soil moisture replenishment. Cold-season precipitation century with no evidence of any long-term trend. Lower deficiencies are also influenced by persistent mid- levels coincided with the droughts of the 1930s, early tropospheric circulation patterns that involve anomalous 1960s, and the most recent 1999-2001 dry period. Over the ridging over the affected area (e.g., Shabbar et al., 1997). Prairies, the numbers and water levels of wetlands have shown no clear trend over the last 40 to 50 years (Conly Reasons for the persistence of anomalous circulation and van der Kamp, 2001). patterns that lead to drought are not entirely understood but are likely related to surface boundary conditions such as Large-Scale Causes snow and ice cover, vegetation, soil moisture, and sea- Although considerable research has been carried out on the surface temperatures (SSTs) that are known to force the subject of droughts, there is still no comprehensive theory climate system through variations in their optical and that explains the physical mechanisms of their formation, thermal properties (e.g. Maybank et al., 1995). The forcing maintenance, and termination. Droughts are caused by factors directly influence local to regional atmospheric flow disruptions to an expected precipitation pattern and can be which in turn can affect large-scale circulation over other intensified by anomalously high temperatures. The major areas of the globe (known as teleconnections). Several factor in the onset and perpetuation of drought involves studies have found relationships between various anomalous circulation patterns in the upper atmosphere. teleconnection indices and climate anomalies over North Few studies have examined atmospheric causes of America. For Canada, significant relationships between El Canadian droughts with the majority focusing on western Niño/Southern Oscillation (ENSO) and winter/spring regions of the country. Over the agricultural region of the temperature and precipitation have been determined Canadian Prairies, growing season (May to August) (Shabbar and Khandekar, 1996, Shabbar et al., 1997). extended dry periods were associated with a persistent Linkages between Canadian climate and teleconnections upper-atmospheric circulation pattern that includes a large- such as the Pacific Decadal Oscillation (PDO) and the North

-81- CMOS Bulletin SCMO Vol.36, No.3 Atlantic Oscillation (NAO) are also evident, but mainly favour the formation of anticyclonic pressure conditions, during the winter season (e.g., Bonsal et al., 2001a). thus perpetuating dry conditions for a season or longer. Teleconnections during summer are generally least robust Fennessy and Sud (1983) suggested that large-scale, as compared to winter. However, some association prolonged droughts may be partially maintained by the between North Pacific SST anomalies and atmospheric feedback influence of soil moisture on rainfall. Further ridging leading to growing season extended dry spells over substantiation of this hypothesis was provided by Oglesby the Prairies has been identified (Bonsal et al., 1993). In and Erickson (1989) who found that a reduction of soil addition, a study by Shabbar and Skinner (2004) indicated moisture in the U.S. Great Plains leads to higher surface that ENSO, the PDO, and their interrelationship play a temperatures, lower surface pressures, increased upper- significant role in summer moisture availability in Canada. atmospheric ridging, and a northward shift of the jet stream. In particular, it was determined that El Niño events are They therefore concluded that soil moisture played an associated with a summer moisture deficit in the western important role in prolonging and amplifying summer drought. two-thirds of Canada while La Niñas produced an An extension of this study found that whereas late winter – abundance of summer moisture in extreme western early spring soil moisture had little effect beyond a couple Canada. The considerable lag between summer moisture of weeks, a late spring anomaly could have a significant and large-scale SSTs provides a basis for developing long- effect on the ensuing summer (Oglesby, 1991). While these range forecasting of drought conditions in Canada. Note studies have focussed on the U.S. Great Plains, there is that analyses of the large-scale atmospheric circulation good reason to believe that they also apply to the Canadian patterns and teleconnections associated with the Canadian Prairies. droughts during 2001 and 2002 revealed that the circulation was markedly different from that associated with previous Future Droughts severe droughts over western Canada. Moreover, the All Global Climate Models (GCMs) are projecting future evolution and persistence of these droughts had no clear increases of summer continental interior drying and relationships with teleconnection patterns that have been associated risk of droughts. The increased drought risk is shown to influence past climate extremes over the country. ascribed to a combination of increased temperature and Results suggested that the dry conditions were related to a potential evapotranspiration not being balanced by northward extension of positive 500 hPa height anomalies precipitation (Meehl and Stocker, 2007). A first-order over the continental U.S. that persisted for several seasons. assessment of future drought occurrence over southern However, further research into this possible link is required Canada by Bonsal and Regier (2006) (Figure 2) indicates (Bonsal and Wheaton, 2005). small positive changes in SPI over all of southern Canada for the majority of climate-change scenarios. This is Several studies have examined large-scale causes of major attributable to the fact that future annual precipitation is drought episodes over the mid-western U.S. These projected to increase over most of southern Canada. The investigations have highlighted a number of potentially PDSI findings, however, reveal dramatic increases to the important factors contributing to the dry conditions including potential for future droughts particularly, in the warm-dry extra-tropical and tropical Pacific SST anomalies, soil scenario. Differences between the SPI and PDSI can be moisture, changes in storm tracks and links with adjacent attributed to the PDSI’s inclusion of temperature in its Pacific and Atlantic anticyclones (e.g., Trenberth et al., calculation. All scenarios project considerable increases to 1988; Ting and Wang, 1997; Schubert et al., 2004a). In temperature and in general, small increases to precipitation addition, the recent 1998 to 2002 extreme drought over over southern Canada. The higher temperatures translate most of the U.S. and the 1930s dust bowl over the U.S. into greater evaporation and more severe droughts. Since Great Plains have been attributed, in part, to La Niña-like the SPI only takes into account precipitation, droughts using conditions in the tropical Pacific (Hoerling and Kumar, 2003; this index are not projected to become more severe in the Schubert et al., 2004b). Although a few of these future. It should also be pointed out that considerable relationships likely apply to Canada, it appears that some of uncertainty exists with respect to future precipitation, the identified causes differ from those associated with particularly on a regional and intra-seasonal basis. western Canadian droughts. This signifies that further Furthermore, relatively little is known regarding changes to research is required to determine the large-scale causes of large-scale circulation and since these patterns have a extended dry periods over interior North America including significant impact on temperature and precipitation over possible differences between northern and southern Canada, the occurrence of future drought remains a huge portions of these mid-continental regions. knowledge gap.

Droughts tend to persist in that warm, dry springs are Monitoring, Modelling, and Prediction followed by hot, dry summers. There also appears to be a Numerous indices that are measures of drought severity are tendency for warm summers to follow other warm summers, used for the monitoring and modelling of drought. These and so on. Reasons for this are likely related to feedback range from simple approaches that only consider processes that enhance or prolong drought situations (e.g., precipitation (e.g., SPI), to more complex indices that soil moisture anomalies) (Maybank et al., 1995). The incorporate a water balance approach using precipitation, concept that drought begets drought was first proposed by potential evapotranspiration, antecedent soil moisture and Namias (1960) who suggested that dry conditions in spring runoff (e.g., PDSI). Various indices have also been used to

CMOS Bulletin SCMO Vol.36, No.3 -82- monitor and model soil moisture changes from daily three-month deterministic temperature and precipitation precipitation and actual evapotranspiration. A problem with forecasts for the ensuing 1-3, 2-4, 4-6, 7-9, and 10-12 these more complex indices is that evapotranspiration is month periods. The 1-3 and 2-4 month forecasts are based difficult to compute since it relies on meteorological upon an ensemble of 40 model runs using the Climate measurements that are generally not readily available (net version of the Global Environmental Multiscale model radiation, vapour pressure deficit, wind speed). The high (GEM-CLIM), the second generation of the Atmospheric spatial variability of summer convective rainfall and General Circulation Model (AGCM2), the third generation of difficulties in modelling snowmelt and blowing snow also the Atmospheric General Circulation Model (AGCM3), and hinder regional-scale moisture modelling (Maybank et al., the Spectral aux éléments finis (SEF) model. Probabilistic 1995). forecasts, which give estimates of the probability that the seasonal mean will be above, near, or below normal are Real-time reports of lake and reservoir levels, stream flows, also provided for the 1-3 and 2-4 month periods. Forecasts snowpack, water-supply volume forecasts, dugout water for the 4-6, 7-9 and 10-12 month periods are produced with levels (for the Prairies), precipitation, and various drought a statistical method and do not use ensemble forecast indices are currently used for drought monitoring in Canada. technique. The forecasts are updated on a monthly basis. The status of these water supplies is critical for activities such as irrigation, water apportionment, storage, flood Adaptation forecasting, hydro-electric power generation, navigation, Adaptation involves adjusting to climate change, variability, fisheries, and wetland habitat. In the Canadian Prairies, and extremes in order to avoid or alleviate negative impacts provincial water resource agencies have been publishing and benefit from opportunities. Drought adaptations include monthly reports of stream, lake, reservoir, and groundwater short to long-term actions, programs, and polices levels since the late 1970s. For Canada, real-time implemented both during, and in advance of the drought to information on pasture conditions, on-farm surface water help reduce risks to human life, property, and productive supplies, and several drought indices are provided in capacity (Wilhite, 2000). Canadians have a great deal of Agriculture and Agri-Food Canada’s Drought Watch web experience in adapting to droughts; however, their site (http://www.agr.gc.ca/pfra/drought/default.htm). The site strategies vary by sector and location. Areas with greater monitors the risk and status of drought over major drought risks are often better prepared to deal with dry agricultural regions of the country by providing national and conditions. Drought adaptation decisions are made at a regional maps of temperature, precipitation, SPI, PDSI, and variety of levels ranging from individuals, to groups and soil moisture for various time periods. It also promotes institutions, to local and national governments. There are practices to reduce drought vulnerability. Recently, the various adaptation processes or strategies including sharing North American Drought Monitor and/or bearing the loss, modifying drought effects, research, education, behavioral changes, and avoidance (Burton et al., 1993). Adaptive drought measures include soil and http://www.ncdc.noaa.gov/oa/climate/monitoring/ water conservation, improved irrigation, the construction of drought/nadm/ infrastructure (wells, pipelines, reservoirs), and the exploration of groundwater supplies. The usefulness of was established as a cooperative effort among drought these strategies varies with location, sector, and nature and experts in Canada, Mexico, and the U.S. to monitor drought timing of the drought. Better management responses may across the continent on an ongoing basis. The Monitor be made with improved drought monitoring and advanced provides continental maps that aid in assessing and prediction. Adjustments that occur after the drought are communicating the state of drought over North America on generally less effective compared to planned anticipatory a weekly basis. The maps are based on a process that adaptation. synthesizes multiple indices, outlooks and local impacts, into an assessment that best represents current drought Drought adaptation research and planning strategies are in their early stages although risk management plans for conditions. drought prone regions of the country have been established (e.g., Agriculture Drought Risk Management Plan for Satellite and radar measurements can potentially provide Alberta). Many adaptive strategies have been devised and solutions to the spatial-scale problems associated with tested for their effectiveness in reducing drought impacts drought monitoring and modelling. The Meteorological (Maybank et al., 1995). However, intense, large-area Service of Canada currently uses Special Sensor droughts that persist for several years still result in severe Microwave Imager (SSMI) to produce snow water hardship to even those regions used to coping with equivalent maps for the Prairie provinces that are made droughts. An improved capability to estimate the numerous available to water resource agencies. impacts associated with drought is required for enhanced adaptation. Future national, provincial, and municipal level Drought prediction involves anticipating climatic anomalies coordinated and proactive drought planning is also needed that produce unusually dry conditions for an extended since vulnerability to future droughts could be exacerbated period of time. However at present, there is no completely by increasing development, and the increased risk of satisfactory method that can routinely predict Canadian drought that is projected to occur over most mid-latitude climate over the month to season time frame required for continental interiors as a result of climate change. drought analysis. Environment Canada currently produces

-83- CMOS Bulletin SCMO Vol.36, No.3 SPI PDSI

Figure 2: Projected average changes in SPI and PDSI values over southern Canada based on a) average, b) warm and dry, c) warm and wet, and d) cool and wet climate-change scenarios. The scenarios are based upon 30-year projected changes in temperature and precipitation for the period 2041-2070 (relative to the 1961-1990 baseline) using the A2 and B2 emission scenarios from multiple GCMs. Maps provide agricultural year (September to August) SPI and PDSI changes (from Bonsal and Regier, 2006).

Knowledge Gaps and Program Needs drought, particularly during the summer season; There are several gaps in the knowledge of droughts that b) The impacts of soil moisture anomalies on the limit our ability to understand their occurrence, perpetuation and migration of drought; monitor/model their status, and adapt to their negative c) The physical causes of multi-year droughts and effects. The following identifies major research and program their recurrence on decadal time scales; needs regarding droughts in Canada. d) The atmospheric circulation patterns associated with unusually large spatial-scale droughts (e.g., 1) Drought Occurrence the 2001 drought over most of southern Canada); A better understanding of the physical causes and e) The atmospheric conditions responsible for the characteristics of past droughts including their spatial and termination of a drought including aspects such as temporal variability is required. This will provide improved convective rainfall, precipitation trigger insight toward short-term (seasonal to annual) and long- mechanisms and moisture sources. term (decade to century) projections of future droughts in Canada. In particular, we require: # Better knowledge regarding the occurrence of future droughts in terms of likely areas to be affected and potential # Improved knowledge of drought trends and variability changes to their frequency, duration, and severity. This prior to the instrumental record. This requires more research requires: into reliable proxy indicators to reconstruct drought occurrence over various regions of Canada for the last few a) More reliable future climate simulations hundred years. (particularly precipitation) from Global and Regional Climate Models; # An improved understanding of the physical causes of b) Improved downscaling methods for the drought initiation, persistence and termination during the application of climate model data to the appropriate last few hundred years. This includes: spatial and temporal scales; c) Knowledge of future changes to large-scale a) The role of large-scale atmospheric and oceanic circulation patterns and oscillations such as ENSO, oscillations in the initiation and persistence of PDO and the NAO. anomalous circulation patterns responsible for

CMOS Bulletin SCMO Vol.36, No.3 -84- 2) Monitoring, Modelling, and Prediction losses. The ability to more accurately predict drought onset, intensity, and termination requires improvements in the # More research into the understanding and modelling of modelling and monitoring of current drought conditions, as drought adaptation measures, including their effectiveness, well as, better seasonal climate forecasts. The following are practicality, costs and benefits. needed to improve our capability to monitor, model and predict Canadian droughts: # Improved knowledge regarding adaptation to prolonged droughts including those that may result due to climate # Development of a total water supply database including change. for example, improved data of stream flow records, wetland numbers and groundwater supplies. # Better abilities to adequately assess the socio-economic consequences of alternative drought adaptation strategies. # Development of an index or combination of indices to monitor past and near real-time drought conditions and to References aid in the recognition of drought sufficiently in advance. Standard indices would allow for national-scale evaluations Bonsal, B.R., Chakravarti, A.K. and Lawford, R.G., of drought. “Teleconnections between North Pacific SST anomalies and growing season extended dry spells on the Canadian Prairies”, International Journal of Climatology, Vol. 13, pp. 865-878 (1993). # A better understanding of the amount and distribution of groundwater resources including the linkages to climate and Bonsal, B.R., Shabbar, A. and Higuchi, K., “Impacts of low surface water supply. frequency variability modes on Canadian winter temperature”, International Journal of Climatology, Vol. 21, pp. 95-108 (2001a). # Development of better methodologies to incorporate remote sensing and ground-level radar for drought Bonsal, B.R. and Regier, M., “The 2001 and 2002 Canadian monitoring and management (to augment the climate Drought: Historical Context and Potential Future Occurrence”, station network). The geospatial and temporal capacity of Environment Canada Water Science and Technology Directorate satellite imagery offers many opportunities for advanced internal report No. AE1-TN-06-002, 58 pp (2006). monitoring capabilities. Bonsal, B.R. and Regier, M., “Historical comparison of the 2001/2002 drought in the Canadian Prairies”, Climate Research, # The incorporation of existing Geographical Information Vol. 33, pp. 229-242 (2007). System (GIS) techniques to provide better spatial representations of drought. For example, the migration Bonsal, B.R. and Wheaton, E.E., “Atmospheric circulation patterns of drought and its associated synoptic circulation comparisons between the 2001 and 2002 and the 1961 and 1988 patterns could be tracked on a variety of temporal scales. Canadian Prairie droughts”, Atmosphere-Ocean, Vol. 43, pp. 163- 172 (2005). # Better hydrologic modelling techniques and in particular, improved methodologies to estimate evapotranspiration. Bonsal, B.R., Zhang, X., Vincent, L.A. and Hogg, W.D, “Characteristics of daily and extreme temperatures over Canada”, Journal of Climate, Vol. 14, pp. 1959-1976 (2001b). # Improved integration of Global and Regional Climate Models with distributed water balance models in order to Brown, R.D., “Northern Hemisphere snow cover variability and model future drought conditions. change, 1915-97”, Journal of Climate, Vol. 13, pp. 2339-2355 (2000). # More reliable short-term (seasonal) forecasts of temperature and precipitation at the appropriate spatial Burton, I., Kates, R.W. and White, G.F., “The Environment as a scales to aid in the prediction of drought onset, intensity, Hazard”, The Guilford Press, New York, NY, 290 pp (1993). persistence and termination. Chakaravarti, A.K., “Precipitation deficiency patterns in the Prairie Provinces of Canada”, Prairie Forum – The Journal of the 3) Impacts and Adaptation Canadian Plains Research Centre, Vol. 1, pp. 95-110 (1976). Droughts are certain to recur in the future. As a result, more effective short and long-term adaptation strategies are Chipanshi, A.C., Findlater, K.M., Hadwen, T. and O’Brien, E.G. required to defend against these future droughts including “Analysis of consecutive droughts on the Canadian Prairies”, improved technological, monitoring and predictive Climate Research, Vol. 30, pp. 175-187 (2006). capabilities. Additional requirements include: Conly, F.M. and van der Kamp, G., “Monitoring the hydrology of # More rapid updates of potential drought conditions to Canadian Prairie wetlands to detect the effects of climate change activate drought response and adaptation options. and land use changes”, Journal of Environmental Monitoring and Assessment, Vol. 67, pp. 195-215 (2001).

# The identification of ecosystem and economic thresholds Fennessy, M.J. and Sud, Y., “A study of the influence of soil to determine at what point during a drought adaptation moisture on future precipitation”, Technical Memo 85042, NASA options need to be activated to avoid serious or irreversible Goddard Space Flight Center, Greenbelt, MD, 125 pp. (1983).

-85- CMOS Bulletin SCMO Vol.36, No.3 Hoerling M. and Kumar, A., “The perfect ocean for drought”, Schubert, S.D., Suarez, M.J., Pegion, P.J., Koster, R.D., and Science, Vol. 299, pp. 691-694 (2003). Bacmeister, J.T., “On the cause of the 1930s dust bowl”, Science, Vol. 303, pp. 1855-1859 (2004b). Maybank, J., Bonsal, B.R., Jones, K., Lawford, R.G., O’Brien, E.G., Ripley, E.A. and Wheaton, E., “Drought as a natural Skinner, W.R.,”Update on western Canada drought: The Palmer disaster”, Atmosphere-Ocean, Vol. 33, pp. 195-222 (1995). Drought Severity Index”, Unpublished report (2002).

Meehl, G.A. and Stocker, T.F., “Global Climate Projections”, Shabbar, A., Bonsal, B.R. and Khandekar, M., “Canadian Chapter 10 in “Climate Change 2007: The Physical Science precipitation patterns associated with the Southern Oscillation”, Basis”, Intergovernmental Panel on Climate Change, Cambridge Journal of Climate, Vol. 10, pp. 3016-3027 (1997). University Press, Cambridge, UK, pp 747-845 (2007). Shabbar, A. and Khandekar, M., “The impact of El Niño-Southern McKay, G.A., Godwin, R.B. and Maybank, J., “Drought and Oscillation on the temperature field over Canada”, Atmosphere- hydrological drought research in Canada”. Canadian Water Ocean, Vol. 34, pp. 401-416 (1996). Resources Journal, Vol. 14, pp. 71-84 (1989). Shabbar, A. and Skinner, W., “Summer drought patterns in Canada MSC Drought Study Group, “An Applied Climatology of Drought in and the relationships to global sea surface temperatures”, Journal the Canadian Prairie Provinces”, Report No. 86-4, Canadian of Climate, Vol. 17, pp. 2866-2880 (2004). Climate Centre, Downsview, ON, 197 pp. (1986). Ting, M. and Wang, H., “Summertime U.S. precipitation variability Namias, J., “Factors in the initiation, perpetuation, and termination and its relation to Pacific sea surface temperature”, Journal of of drought”, In: Publication 51, Association of Scientific Hydrology, Climate, Vol. 10, pp. 1853-1873 (1997). International Union of Geodesy and Geophysics, pp. 109-111 (1960). Trenberth, K.E., Branstator, G.W. and Arkin, P.A., “Origins of the 1988 North American drought”, Science, Vol. 242, pp. 1640-1645 Oglesby, R.J. and Erickson, D.E., “Soil moisture and the (1988). persistence of North American drought”, Journal of Climate, Vol. 2, pp. 1362-1380 (1989). Wheaton, E., Wittrock, V., Kulshreshtha, S., Koshida, G., Grant, C., Chipanshi, A. and Bonsal, B.R., “Lessons Learned from the Oglesby, R.J., “Springtime soil moisture, natural climatic variability, Canadian Drought Years of 2001 and 2002: Synthesis Report”, and North American drought as simulated by the NCAR Agriculture and Agri-Food Canada, Saskatchewan Research Community Climate Model 1”, Journal of Climate, Vol. 4, pp. 890- Council Publication No. 11602-46E03, Saskatoon, SK, (2005). 897 (1991). Wilhite, D., “Drought as a Natural Hazard: Concepts and Phillips, D.W., “The Climates of Canada”, Environment Canada, Definitions”, Part I in: Drought, A Global Assessment, Vol. 1., Ottawa, ON, 176 pp. (1990). Routledge Press, New York, NY (2000).

Phillips, D.W., “The top ten Canadian weather stories for 2001”, Zhang, X., Harvey, K.D., Hogg, W.D. and Yuzyk, T.R., “Trends in CMOS Bulletin SCMO , Vol. 30, pp. 19-23 (2002). Canadian streamflow”, Water Resources Research, Vol. 37, pp. 987-998 (2001). Sauchyn, D.J. and Skinner, W.R., “A proxy record of drought severity for the southwestern Canadian plains”, Canadian Water Zhang, X., Vincent, L.A., Hogg, W.D. and Niitsoo, A., “Temperature Resources Journal, Vol. 26, pp. 253-272 (2001). and precipitation trends in Canada during the 20th century”, Atmosphere-Ocean, Vol. 38, pp. 395-429 (2000). Schubert, S.D., Suarez, M.J., Pegion, P.J., Koster, R.D., and Bacmeister, J.T., “Causes of long-term drought in the U.S. Great Plains”, Journal of Climate, Vol. 17, pp. 485-503 (2004a).

URGENT - URGENT - URGENT - URGENT

Prochain numéro du CMOS Bulletin SCMO

Le prochain numéro du CMOS Bulletin SCMO paraîtra en août 2008. Prière de nous faire parvenir avant le 4 juillet 2008 vos articles, notes, rapports d’atelier ou nouvelles à l’adresse indiquée à la page ii. Nous avons un besoin URGENT de vos contributions écrites.

CMOS Bulletin SCMO Vol.36, No.3 -86- A Drought Research Initiative for the Canadian Prairies

Ronald Stewart1, John Pomeroy2 and Rick Lawford3

Résumé: Les Prairies sont souvent exposées à la sécheresse et quelques fois de façon catastrophique. La plus récente sécheresse a eu lieu au cours de l’épisode 1999 à 2005, et elle a produite des conditions les plus sèches du dossier historique. Pour traiter de telles sécheresses, un réseau de recherche du nom de DRI (Drought Research Initiative – Initiative de recherche sur la sécheresse) a été formé. Le centre d’intérêt particulier du réseau DRI consiste à mieux comprendre la façon dont les sécheresses surviennent, évoluent et prennent fin afin de comprendre leur structure interne et de contribuer à une meilleure prédiction de tels événements. Afin de réaliser cet objectif, on considère la sécheresse sous différents points de vue en tenant compte de l’atmosphère, du sol incluant les processus de la végétation et du sous-sol. Cette sécheresse fut exceptionnelle de telle sorte que son processus à grande échelle a été assez variable dans le temps. De plus, on a noté que les nuages étaient fréquents et qu’à travers les Prairies il est arrivé de voir quelques fois, de façon simultanée, des records de grosses précipitations dans certaines régions. Pourtant, cette sécheresse a produit quelques-unes des plus grandes diminutions de l’humidité enregistrée en profondeur et a contribué à une diminution majeure de l’écoulement des rivières. De plus, la communauté de recherche du réseau DRI, à travers le Canada, travaille étroitement avec plusieurs partenaires concernés par la sécheresse et qui ont tous intérêt à faire face à de tels événements, dans le futur. Cet article résume l’Initiative de recherche sur la sécheresse en faisant valoir les caractéristiques de la récente épisode de sécheresse, ainsi que les objectifs du réseau DRI, les enjeux scientifiques clés, les récents progrès et les plans futurs.

1. Background The occurrence of drought is a ubiquitous feature of the short growing season; local surface runoff is primarily global water cycle. Such an extreme does not necessarily produced by snowmelt water; and the landscape has lead to an overall change in the magnitude of the global substantial water storage potential in the poorly drained, water cycle but it of course affects the regional cycling of post-glacial topography; and aquifers are overlain by water. Droughts are recurring aspects of weather and impermeable glacial till, but there are also important climate extremes, as are floods and tornadoes, but they permeable aquifers in surficial deposits that have differ substantially since they have long durations and lack substantive water exchange with the surface. easily identified onsets and terminations. One example of Prairie drought is the recent multi-year Drought can be considered as an anomaly within the event that began in 1999 with cessation of its atmospheric atmospheric, surface and sub-surface cycling of water and component in 2004/2005 and many of its hydrological energy (Figure 1). This anomaly is initiated through large components in 2005. This event produced the worst drought scale atmospheric processes and is enhanced and for at least a hundred years in parts of the Canadian maintained through regional to local atmospheric, surface Prairies. An illustration of the severity is shown in Figure 2 hydrology, land surface and groundwater feedbacks that for the 2000/01 agricultural year. This multi-year drought operate at various time scales over the growing season and led to major impacts across the Prairies as discussed by during dormant snow-free and snow-covered periods. Lawford et al. (2008).

Drought is a relatively common feature of the North Despite enormous economic, environmental, and societal American and Canadian climate system and all regions of impacts of drought, there has never been a coordinated and the continent are affected from time-to-time. However, it integrated drought research program in Canada. Given the tends to be most common and severe over the central importance of this extreme form of weather, it is critical that regions of the continent. The agricultural areas of the it be studied appropriately with the hope being that its Canadian Prairies are therefore prone to drought. occurrence and nature can be better anticipated on short and long term scales. Droughts in the Canadian Prairies are distinctive in North America. The large scale atmospheric circulations are The purpose of this article is to provide a brief overview of influenced by blocking from major orographic features to the a Drought Research Initiative (DRI). west and long distances from all warm ocean-derived atmospheric water sources. Growing season precipitation is generated by a highly complex combination of frontal and convective systems. Seasonality is severe and characterized by a relatively long snow-covered season and

1 McGill University, Montréal, QC 3 University of Manitoba, Winnipeg, MA

2 University of Saskatchewan, Saskatoon, SK

-87- CMOS Bulletin SCMO Vol.36, No.3 Figure 1. Water and energy cycling associated within Canadian Prairie drought.

2. Objectives To begin to address these issues related to drought, the 1) To quantify the physical features of this recent drought; Drought Research Initiative, DRI was established. DRI is a research network largely funded by the Canadian 2) To improve the understanding of the processes and Foundation for Climate and Atmospheric Sciences feedbacks governing the formation, evolution, cessation and (CFCAS). structure of the drought;

The overall objective of DRI is "to better understand the 3) To assess and reduce uncertainties in the prediction of physical characteristics of and processes influencing drought and its structure; Canadian Prairie droughts, and to contribute to their better prediction, through a focus on the recent severe drought 4) To compare the similarities and differences of the recent that began in 1999 and largely ended in 2005". drought to previous droughts over this region and those in other regions, in the context of climate variability and To address this overall objective, the Network is focused on change; and five research objectives: 5) To apply our progress to address critical issues of importance to society.

CMOS Bulletin SCMO Vol.36, No.3 -88- Figure 2. Example of the precipitation deviation from normal in the early drought period, September 1, 2000 - August 31, 2001. Courtesy of Prairie Farm Rehabilitation Administration.

DRI brings together many university and federal/provincial successfully for IPY (International Polar Year) and for government researchers to address this issue with expertise MAGS (Mackenzie GEWEX Study) which used it to encompassing the atmospheric, hydrologic, land surface examine a year with record low discharge from the and predictive aspects of droughts at a variety of spatial and Mackenzie River (Stewart et al., 2002). The assumption is temporal scales. As shown in Table 1, DRI is overseen by that, in general, it is often only feasible to bring together for a Board of Directors, its scientific effort is driven by a a particular event or period the necessary full complement Science Committee and its researchers have either CFCAS of observations, models and scientific research capability to funding (investigators) or are carrying out contributing examine phenomena such as drought. In the case of DRI, research through other means (collaborators). It should be one can even consider the focus on 1999-2005 drought as noted that only Objectives 1-3 were able to be funded being a 'pilot' project for drought research including strong directly through CFCAS; objectives 4-5 are being addressed interactions with those affected by it. using other funds and linkages to the extent possible. Lawford et al. (2008) discusses the critical DRI Partners From a longer-term perspective, it is envisioned that this 5- Advisory Committee that gives guidance directly to the year Network represents an essential step towards our Board of Directors. ultimate goals:

To make progress on this critical issue over the 5-year # To better predict droughts over Canada, their period of a CFCAS Network, the project focuses, although detailed structure and their impacts with increasing not exclusively, on the recent drought over the Prairies. This confidence; drought was by far the best observed and modelled one over this region and its impacts are still fresh in people's # To better assess whether there will be a ‘drying memory. This strategy, a focus on physical processes and of the continental interior’ in the future. particular time periods or events, is a common one although not so for drought. For example, it has been applied

-89- CMOS Bulletin SCMO Vol.36, No.3 These two overarching issues directly address major issues also an important long-term indicator of drought. An for society. There needs to be better guidance on the additional issue for the Prairies is that they are covered by likelihood of drought including its detailed and varying undulating or hummocky terrains with numerous structure. Furthermore, there is tremendous concern that topographic depressions. The majority of individual our changing climate will lead to more periods of drought in depressions are too small to be captured in 1:50,000 the future over the interior of North America (see for topographic maps, but collectively they represent an example Lemmen et al., 2008). The implications of this are enormous capacity to store surface water without allowing so far reaching that this has to be our ultimate objective; to it to be drained to streams. Proper characterization of better assess the likelihood of such a scenario. With this depression storage is essential for understanding the Network, we will be moving systematically towards these hydrology of the prairie region. It is not easy to acquire the two objectives. needed observational information. First, there is operational instrumentation across the region that in some cases is 3. Strategy arranged as networks. This includes weather stations, As outlined in Section 2 above, DRI will achieve its overall radars, lightning detectors, stream gauges, lake levels, objective through a focus on five complementary, cross- vegetation assessment, crop yields, snow information, soil cutting research objectives or themes. The five themes moisture and ground water. In general, though, there are represent a logical sequence for such a network by relatively few of these sites. Others though have used some including the quantification, understanding and better of these data in order to make gridded products that are prediction of a particular drought and the subsequent extremely useful for characterizing drought. One example comparisons of our findings with other droughts as well as of this is the CANGRID precipitation information based on the implications for society. station information that has been homogenized and adjusted for all known measurement errors (Mekis and 3.1 Theme 1 Hogg, 1999; Vincent and Gullett, 1999). Second, satellite- The first step towards the better understanding of the 1999- based observations are extremely important and they 2005 drought involves the quantification of its atmospheric, provide critical information on a host of variables in the hydrologic and land-surface physical features at a variety of atmosphere, at the surface and even sub-surface. Variables spatial and temporal scales. As the drought is being range from clouds and precipitation, to snowcover and quantified, this information forms much of the basis of vegetation, down to sub-surface water storage. Not all of Themes 2 through 5. the products are adequate however and many have limitations on temporal and spatial resolution. Third, there Theme 1 is being realized through three focussed are a few locations (Figure 3) at which a large number of questions: detailed measurements are carried out. This includes BERMS (Boreal Ecosystem Research and Monitoring Sites) # What variables are required to quantify the in northern Saskatchewan, the St. Denis Wildlife area near characteristics of this recent drought? Saskatoon, and the Assiniboine Delta Aquifer in western Manitoba. The types of data at these sites is not consistent # What data sources and model outputs are though with some, for example, having a number of available for quantification of these parameters? atmospheric-related measurements and others focused on surface and sub-surface variables. Fourth, some individual # How do we characterize and “close the budgets” researchers have acquired their own unique field of water and energy over the Prairies? measurements. There is a wide variety of this information and much of it can be accessed through contacting the The quantification of drought necessarily means that many researcher directly. fluxes and reservoirs need to be addressed. This is schematically illustrated in Figure 1. In particular, variables Model products are being used to fill gaps in the characterizing atmospheric, surface and sub-surface states observational record. These data sets include GEM (Global are required. A three-dimensional assessment of the Environmental Multiscale model), ECMWF (European atmosphere during drought over various temporal scales Centre for Medium range Weather Forecasting), CRCM requires, for example, knowledge of temperature, humidity, (Canadian Regional Climate Model, and CCCMA (Canadian pressure, wind, clouds and precipitation amount. The Centre for Climate Modelling and Analysis) to name a few. transfer of heat and moisture between the surface and Each of these models has its own spatial and temporal atmosphere must also be assessed. At the surface, the resolution, from coarse spatial scales (of order a few spatial and temporal characteristics of vegetative state (in hundred kilometres) to shorter mesoscale ones (a few tens terms of water stress) for major vegetation types (crops and of kilometres). boreal zones), soil moisture, stream network, river flows, lake levels, wetlands and depression storage are required to assess when and where drought is occurring. Sea surface temperatures are also needed to characterize global connections with prairie drought. The spatial characteristics of groundwater and sub-surface moisture are

CMOS Bulletin SCMO Vol.36, No.3 -90- a)

Synoptic Processes Planetary Scale Atmospheric Mesoscale Processes Teleconnections

Deep Convective Processes

Streamflow Processes Vertical Scale Vertical Land Surface Hydrological/Biophysical Processes

Groundwater Processes

1m 100 m 1 km 10 km 100 km 1000s km Horizontal Scale

b)

Synoptic Processes Planetary Scale Atmospheric Mesoscale Teleconnections Processes

Deep Convective Processes

Streamflow Vertical Scale Vertical Land Surface Hydrological/Biophysical Processes

Groundwater Processes

1min 1 hr 1 day 1 week 1 season 1 year 10 years Temporal Scale

Figure 3: (a) Spatial and (b) temporal scales of processes associated with Prairie droughts. Note that 'mesoscale' refers to the atmospheric processes including the precipitation associated with frontal systems.

-91- CMOS Bulletin SCMO Vol.36, No.3 Hydrological and surface models are essential to DRI. Ones It also needs to be recognized that there are many ways in being used and/or improved extensively include MESH which drought, basically a sustained precipitation deficit, (MEC [Modélisation Environmentale Communautaire] can be prolonged. Some of the ways to reduce precipitation Surface and Hydrology model interfacing with the GEM include, for example, large scale circulation anomalies, lack model), VIC (Variable Infiltration Capacity model), CRHM of moisture advected into a region, reduction of local (Cold Regions Hydrological Model), SABAE-HW (Soil moisture supplies, reduction of vegetation for Atmosphere Boundary, Accurate Evaluations of Heat and evapotranspiration, reduction of sub-surface water supply Water), and CLASS (Canadian Land Surface Scheme). for evaporation, production of virga as opposed to This effort includes coupling of surface and sub-surface precipitation, and possibly even the role of aerosols in the water conditions. dusty environment. Over the 5+ years of the drought, it is anticipated that all these, and numerous other factors, were Furthermore, the fluxes and budgets of water and energy operating at various times and locations to suppress the through and within the Prairies during drought must be production of substantial precipitation. addressed. This needs to be accomplished through the analysis of the model and observational products described 3.3 Theme 3 above. This DRI-wide activity is being addressed on scales Given that the 1999-2005 drought and its features are being ranging from long-term values over the entire Prairies to quantified and the fundamental responsible processes smaller scales less than a month over smaller domains that better understood, the next issue is to assess and improve experienced different degrees of impact. predictive techniques for it. The modelling tools used are global and regional climate models (GCMs and RCMs), and 3.2 Theme 2 hydrological models. The hydrological models will be driven Given that the 1999-2005 drought and its features are being by output from the atmospheric models and observations quantified in Theme 1, the next issue is to better understand from research sites and available reanalysis data. the processes and the feedbacks responsible for the Atmospheric modelling spans scales from global to regional drought’s initiation, persistence, and termination. This to watershed scales characteristic of the prairies, while theme is addressing the following questions: hydrological modelling is accomplished using a hierarchy from small scale detailed process models to large scale # What processes were responsible for the onset models run over river basins. The major questions are: of the recent drought? # How well was the current drought predicted # What processes and feedbacks contributed to based on current techniques? the drought’s evolution, persistence, and spatial structure? # To what extent could this prediction be improved through better initialization? # What processes and feedbacks controlled the termination of this drought? # To what extent could this prediction be improved through dynamical downscaling and better Much of DRI's effort is being devoted to better physics? understanding processes and feedbacks. This recognizes that there has been relatively little attention paid to the # What are the appropriate scales and processes mechanisms directly associated with Canadian drought. for prediction of Prairie droughts? Prairie drought processes have characteristic spatial and temporal scales; these are shown in Figure 3. We Use will largely be made of archived information to address hypothesize that atmospheric circulation patterns normally many of these issues. This, for example, is available from induce the onset of drought, but its termination is also linked the Historical Forecast Project that addressed seasonal with land surface and groundwater conditions as well as prediction as well as from operational seasonal forecasts of atmospheric conditions. It is also important to note that land Environment Canada. DRI-related research is focussed on surface hydrological/biophysical processes generally a particular extreme and examines many more variables operate at small spatial scales (and with large inherent than the traditional analysis of temperature and spatial variability) and that there is a hierarchy of precipitation. As discussed under Theme 2, there are many atmospheric process scales. The horizontal scale link ways in which a precipitation deficit arises and these need between land surface and groundwater processes (which to be considered when analyzing prediction results. It is also produce many of the greatest societal impacts) and the expected that experiments will be carried out within DRI for large atmospheric scales is provided by a cascade of assessing the role of soil moisture and perhaps snowcover atmospheric processes and by large basin streamflow. on seasonal prediction. Overall, however, the evolution of drought across different scales is not well understood, and DRI is directly addressing 3.4 Theme 4 this issue. Given that the 1999-2005 drought and its features are being quantified, better understood, and potentially better predicted, the next issue is to compare our progress with

CMOS Bulletin SCMO Vol.36, No.3 -92- droughts occurring at other times and locations. The international and, to a lesser extent, national programs that objectives of Theme 4 will be realized through the following have addressed extremes in the water cycle, DRI fills a research questions: niche because none have specifically focused on the Canadian Prairies. # How do the physical features, processes, and feedbacks of the recent Canadian Prairie drought 4.1 Canadian Linkages compare with a) previous droughts over the On a research level, the Network has already linked several Canadian Prairies, b) Canada-wide droughts, c) US university-based researchers from institutions in Canada as Great Plains droughts, and d) droughts across the well as a number of federal and provincial agencies and world? departments. At the federal level, it has also established collaborative links with several departments including #How does the prediction of the recent drought several components of Environment Canada, Agriculture compare with predictions of other droughts? and Agri-Food Canada, and Natural Resources Canada. DRI has also established close links with Canadian GEO #How does the recent drought compare with past (Global Earth Observations) which is keenly interested in climate variability and projected climate change? drought. DRI is exploring the feasibility for being a test-bed for their pilot projects and long-term plans. Bonsal (2008) identified a number of aspects of the recent drought in relation to others based on circulation patterns DRI has established many linkages at the provincial level. and climate variables. To carry out comprehensive Within the Prairie provinces this includes the Alberta comparisons, however, these analyses have to be extended Agriculture Food and Rural Development; Alberta to include many more variables in the atmosphere and Environment; Manitoba Agriculture, Food and Rural surface. Furthermore, these analyses have to examine the Initiatives; Manitoba Hydro; Manitoba Water Stewardship; internal structure of drought, not just its presence over a Prairie Adaptation Research Collaborative; Saskatchewan particular region. Many droughts, including the one in 1999- Agriculture; Saskatchewan Environment; Saskatchewan 2005, illustrate a complex, dynamic, internal structure of Watershed Authority; and the Saskatchewan Research precipitation and possibly other variables, and the pattern Council. Linkages also include the Prairie Provinces Water changed dramatically during the drought. Such internal Board which considers water flows across provincial features can in some instances be just as important as the boundaries. Links have also been established with general occurrence of drought. Ouranos, the Québec-based initiative concerned with regional climate. 3.5 Theme 5 Theme 5, addressing the societal impacts of drought, can The Network complements previous or current Canadian be broken down into sub-issues which include: research studies. This includes the Mackenzie GEWEX Study (MAGS) that was concerned with Mackenzie basin # What organizations are affected by drought? water and energy cycle issues (Stewart et al., 1998), and Canadian CLIVAR activities that considered climate # What is the nature of the drought’s impact and variability (Derome et al., 2004). DRI also has links with a how can these impacts be alleviated? recently initiated major project on institutional adaptation to climate change, especially water scarcity issues (a # Given the progress being made by DRI, how can collaborative research initiative led by the University of it enable organizations affected by drought impacts Regina). to respond more effectively in the future? Although DRI can link with other Canadian activities, it is A major portion of this effort is described in Lawford et al. unique and fills an important gap. No other activity is (2008). This Theme relies heavily on the active two-way addressing head-on the critical multi-disciplinary drought involvement of the partners. Progress being made by DRI issue that regularly affects the Canadian Prairies. needs to be conveyed to impacted groups and vice versa. In turn, impacted groups through this Theme inform the 4.2 International Linkages Network on critical issues, thresholds and other factors Internationally, DRI has its strongest linkages with the relating to drought, enabling DRI research to adjust and United States. The western portion of the U.S. has also thereby more effectively address the concerns of been experiencing drought conditions for several years and Canadians. there is an increasing sense that a more coordinated effort is needed to address drought (Western Governors’ 4. Linkages Association, 2004). Although there is not yet a comparable Whereas previous studies have focused on individual U.S. Drought Network, strong linkages will be developed features occurring over the Prairies (e.g., large-scale with organizations and individuals studying American circulation, storms, land surface processes, and river flows), droughts. There is, for example, a natural link with the DRI is focussing on the entire ‘system’ during periods of GEWEX America Prediction Project (GAPP where GEWEX extreme dryness. Even though there are several is the Global Energy and Water Cycle Experiment) which in

-93- CMOS Bulletin SCMO Vol.36, No.3 turn is part of CPPA (Climate Prediction Program of the Tremendous progress has been made in examining the Americas). The Network will also interact with the U.S. multitude of processes that operated during the drought. As Drought Mitigation Center, the North American Drought discussed by Bonsal (2008) large scale factors illustrated Monitor, and US CLIVAR activities. consistencies with and differences from those associated with other droughts. However, a host of other atmospheric, DRI is also contributing to and taking advantage of research surface and sub-surface processes including over other parts of the world. For example, both the World evapotranspiration, runoff generation and sub-surface water Climate Research Programme (WCRP) and GEWEX transport are also being studied and these can all affect the through it Coordinated Energy and water cycle Observation nature of drought. All of these issues will be brought Project (CEOP) have recently initiated activities focussing together in a dedicated workshop over the next year to on extremes such as droughts. The work in DRI is ensure that we have collectively addressed this key issue. contributing directly to this and, to some degree, it is This workshop will build on a successful evaporation considered to be a model as to how one addresses workshop that was held in 2007. extremes in a comprehensive manner. There had not been a systematic study before of the Other collaborations will include the International Decade capability of season-prediction systems to anticipate for Predictions in Ungauged Basins 2003-2012 (PUB), a drought. This is currently underway within DRI although, in hydrological decade for improving the understanding and general terms, it appears that key aspects of the drought's prediction of hydrological systems where surface persistence and end were not well-anticipated. In addition, observations are not available for model calibration it is recognized that initialization of these models with (Sivapalan et al., 2003). A particular emphasis of PUB is on surface conditions (soil moisture, snowcover) may well hydrological prediction in semi-arid to sub-humid zones. improve predictive capabilities. Initial experience with these Through exchange of techniques and information, DRI will is anticipated by the end of DRI. Some of this work is being both contribute to and benefit from links to PUB. The carried out in cooperation with GOAPP (Global Ocean- Network will continue links to the International Water Atmosphere Prediction and Predictability, another CFCAS- Management Institute (IWMI) in its studies of drought sponsored Network). A dedicated workshop on this issue assessment and mitigation in southern Asia; the IWMI has been held and this activity will also work with our US currently uses Canadian water assessment methodologies colleagues to ensure as much progress as possible. in some of its hydrology studies Largely through other funding beyond that of CFCAS, a number of researchers are able to carry out comparison http://www.iwmi.cgiar.org/drought assessment/index.asp studies. Bonsal (2008) has noted a number of differences and similarities at large scales for example between this The region has certain similarities to the Prairies in that recent drought and others. Much more needs to be done on snowmelt-fed mountain waters supply much of the water to this issue to apply our detailed knowledge of one drought to a semiarid and sub-humid lowland region. other, less well-understood ones. A dedicated workshop on this issue will be held in the last year of DRI with the intent 5. Progress and Update being to apply our knowledge of the 1999-2005 drought to DRI is midway through its activities. Substantial progress others and to place this drought into a larger context, has been made on many fronts. A few illustrations of including the extent to which it might be the harbinger of progress are mentioned here. future drought over this region.

DRI has brought together an unprecedented amount of As mentioned earlier in this article, strong interactions have information on Canadian drought. On its web site developed with those affected by drought. To move this along as efficiently as possible, the DRI Partners Advisory http://www.drinetwork.ca Committee was established, as discussed by Lawford et al. (2008). This Committee is, for example, organizing, funding one can find many links to various datasets that describe permitted,'a simulation exercise' in which experiences from elements of drought in the atmosphere, at the surface or DRI are presented as if corresponding to a future drought. below the surface. Compiling a Prairie-wide database of Such a learning experience will include both researchers provincial groundwater and water levels is part of this effort. and partners. This collective progress is in itself a major accomplishment. Now the task is to ensure that as much of this information It should also be noted that there are many students as possible is integrated and made even more accessible. working within DRI. These students are enjoying a unique A synthesis article characterizing the 1999-2005 drought is opportunity to contribute to our knowledge of drought as in progress and this will be followed by another one well as to gain a wide perspective on the many issues addressing the water and energy budgets during the linked with this phenomenon. The training of these students drought. In addition, new techniques of archiving and represents an important DRI legacy. making available datasets are being explored as another legacy of DRI.

CMOS Bulletin SCMO Vol.36, No.3 -94- Table 1: DRI Board of Directors, Investigators and Some of the DRI Collaborators. Science Committee members are indicated by * and Theme Leads by “t”. The number of collaborators continues to grow.

Jim Bruce - - Chair

Gary Burke - - Member (Environment Canada)

Board of Harvey Hill - - Member (Agriculture and Agri-Food Canada)

Directors John Pomeroy - - Member (DRI Science Committee Co-Chair)

Ron Stewart - - Member (DRI Science Committee Co-Chair

Tim Aston - - Ex-Officio (CFCAS)

*t Barrie Bonsal (Saskatchewan) * John Pomeroy (Saskatchewan)

Paul Bullock (Manitoba) * Ken Snelgrove (Memorial)

John Gyakum (McGill) * Ron Stewart (McGill)

Investigators *t John Hanesiak (Manitoba) Geoff Strong (Alberta)

*t Masaki Hayashi Garth van der Kamp (Saskatchewan)

Henry Leighton (McGill) *t Elaine Wheaton (Saskatchewan)

*t Charles Lin (McGill) * Al Woodbury (Manitoba)

Al Pietroniro (Saskatchewan)

Aaron Berg (Guelph) Amir Shabbar (EC)

George Boer (EC) Dave Sills (EC)

Daniel Caya (Ouranos) Susan Skone (Calgary)

Jacques Derome (McGill) Craig Smith (EC)

Raoul Granger (EC) * Kit Szeto (EC)

Collaborators Ted Hogg (NRCan) Brenda Toth (EC)

Bob Kochtubajda (EC) Jessica Toyra (EC)

Rick Raddatz (Manitoba) Alex Trishchenko (NRCan)

Hal Rithcie (EC) Anne Walker (EC)

Alonso Rivera (NRCan) Shusen Wang (NRCan)

Dave Sauchyn (Regina)

6. Summary Drought is a huge issue for Canada and DRI is addressing progress contributes directly to these groups and their it in a comprehensive manner for the first time. This guidance influences research plans. collaborative effort is providing a better understanding of the physical features of Prairie drought, the processes DRI is carrying out fundamental research on a climatic controlling it, and the capabilities to predict it. DRI is also extreme but it is also targeted research. Given this reality, comparing the recent drought with other earlier Prairie ones at the conclusion of the Drought Research Initiative in and with those occurring elsewhere; which will lead to a December 2010 we expect to say that: more complete picture of the implications for droughts occurring across Canada. Furthermore, interactions with the “We have greatly increased our understanding of drought user community have been established so that DRI's through a focus on the recent 1999-2005 one over the

-95- CMOS Bulletin SCMO Vol.36, No.3 Prairies and we have applied this to improved prediction. References We have left a legacy of comprehensive datasets, improved observational and modelling techniques, a new generation Bonsal, B., 2008: Droughts in Canada: an overview. CMOS of drought scientists, and a public better educated about Bulletin SCMO (in this issue). drought. We have, in partnership with others in Canada and Derome, J., Boer, G.J., Greatbatch, R.J., and Weaver, A.J., 2004: internationally, developed a plan to improve drought and A workshop on climate variability. CMOS Bulletin, 32, 115-119. water cycle prediction at multiple scales.” Lawford, R., B. Girling, I. Hanuta, H. Hill, T. Hogg, B. Oegema, R. As one means of ensuring that these statements are Keller, and A. Warkenton, 2008: A review of some requirements realized, after each annual workshop, we issue a brief for drought information on the Canadian Prairies. CMOS Bulletin update statement. As of January 2008, DRI's status is: SCMO (in this issue).

"We have continued to add datasets to characterize drought Lemmen, D.S., F.J. Lacroix and E. Bush, editors, 2008: From and to investigate the many factors leading to, sustaining Impacts to Adaptation: Canada in a Changing Climate 2007. Gov. and ending drought. We have developed interactions with of Canada, Ottawa, Ontario, 448 p. other groups examining drought and other extremes in the Mekis, E. and W. Hogg, 1999: Rehabiliation and analysis of United States and elsewhere. Our partners have organized Canadian daily precipitation time series. Atmosphere-Ocean, 37, an advisory group to ensure that there are strong two-way 1999, 53-85. interactions with researchers. We are organizing our synthesis article on drought characterization and we have Sivapalan, M., Takeuchi, K., Franks, S., Gupta, V., Karambiri, H., developed a strategy to assess and to contribute to Lakshmi, V., Liang, X., McDonnell, J., Mendiondo, E., O’Connell, improved predictive capabilities." P., Oki, T., Pomeroy, J.W., Schertzer, D., Uhlenbrook, S. and Zehe, E. 2003. IAHS Decade on Predictions in Ungauged Basins The next two years of DRI will bring our effort together (PUB), 2003-2012: Shaping an exciting future for the hydrological sciences. Hydrological Sciences Journal, 48. 857-880. through collective efforts that directly address our objectives. This period is also being used to establish a DRI Stewart, R.E., H.G. Leighton, P. Marsh, G.W.K. Moore, H. Ritchie, Follow-On. Given its progress, DRI is uniquely positioned to W.R. Rouse, E.D. Soulis, G.S. Strong, R.W. Crawford, and B. address drought anywhere in Canada and indeed it is well- Kochtubajda, 1998: The Mackenzie GEWEX Study: the water and positioned to examine hydrometeorological extremes energy cycles of a major North American river basin. Bull. Amer. anywhere in Canada. Meteor. Soc., 79, 2665-2683.

In summary, a comprehensive study of drought over the Stewart, R.E., N. Bussieres, Z. Cao, H.R. Cho, D.R. Hudak, B. Canadian Prairies is underway. This has brought together Kochtubajda, H. Leighton, P.Y.T. Louie, M.D MacKay, P. Marsh, G.S. Strong, K.K. Szeto and J.E. Burford, 2002: researchers from several disciplines and it is working with Hydrometeorological features of the Mackenzie basin climate partners affected by drought. DRI is making a difference. system during the 1994/1995 water year: a period of record low discharge. Atmos.-Ocean, 40, 257-278. Acknowledgements The support of the Canadian Foundation for Climate and Vincent, L.A., and D. W. Gullett, 1999: Canadian historical and Atmospheric Sciences (CFCAS) is greatly appreciated. homogeneous temperature datasets for climate change analyses. Other agencies supporting DRI directly include Environment Int. J. Climatol., 19, 1375-1388. Canada, Agriculture and Agri-Food Canada, Natural Resources Canada, Prairie Farm Rehabilitation and Western Governors’ Association, 2004: Creating a Drought Early Warning System for the 21st Century Environment, Alberta Agriculture and Food, Saskatchewan . Western Governors’ Assoc., Denver, Co., 16 pp. Research Council, Manitoba Water Stewardship and Manitoba Hydro. The authors would like to thank the entire DRI community for their commitment and enthusiasm. They would in particular like to thank Jim Bruce for chairing the Board of Directors, Rachael Reynen for handling DRI's many financial aspects, Patrice Constanza and Matt Regier for their data management efforts, and Peter Lawford for maintaining our web site.

CMOS Bulletin SCMO Vol.36, No.3 -96- User Expectations for the Drought Research Initiative (DRI)

by Rick Lawford1, Harvey Hill2, Elaine Wheaton3, Irene Hanuta4, Alf Warkentin5 and Bill Girling6

1. Introduction Drought is a major natural hazard that primarily affects Canadian Drought Research Initiative (DRI) emerged. DRI western Canada, but other areas of Canada also are was founded on two objectives, namely: (1) to better vulnerable. Drought impacts are complex due to their understand the physical characteristics of and the dependence on the timing, geographical extent, severity, processes influencing Canadian Prairie droughts, and (2) to and duration of a drought. For example, the agriculture contribute to the better prediction of droughts. The industry is very sensitive to precipitation and temperature objectives, themes, science questions and structure of DRI patterns associated with a drought during the growing are elaborated in Stewart et al. (2008). While three DRI season. The snowpack and the spring precipitation can also themes dealing with research to characterize drought, to be important since the soil moisture reserves supply understand drought processes and to predict drought, were moisture needed by plants. Hydrological droughts are often approved for funding by the Canadian Foundation for initiated by a lack of winter precipitation, especially for Climate and Atmospheric Sciences (CFCAS), the joint watersheds with their headwaters in mountainous areas. application projects with the user community did not garner However, spring and summer precipitation determines if the financial support. In spite of the lack of resources, DRI hydrologic drought becomes more severe or diminishes. scientists have been successful in developing a dialogue Additional complexities in drought assessments occur when with the user community. This paper summarizes some water use patterns are considered. highlights from these discussions.

In order to facilitate the research community interactions Table 1. Largest Hazards in Canada (by damage / loss) with the user community there must be a shared (From Etkin et al., 2004) understanding of the definition of drought. For this reason drought is sometimes classified as agricultural, hydrological, Disaster Year(s) Location Cost1 or socioeconomic, or, for some events, all of the above (Maybank et al., 1995). A definition that is often used Drought 1980 Prairies 5.8 considers drought as “a deficiency of precipitation from expected or “normal” that, when extended over a season or Freezing Rain 1998 Ontario to New 5.4 longer, results in insufficient water to meet the demands by Brunswick agriculture, other sectors and the environment.” This Drought 2001/2002+ National 5.0 definition, while useful in many ways, does not account for factors such as wind, temperature, cloudiness, actual Drought 1988 Prairies 4.1 evaporation, soil moisture and reservoir and groundwater storage, all factors that affect drought impacts on human Drought 1979 Prairies 3.4 activities. Drought 1984 Prairies 1.9

The onset and extent of droughts are not fully appreciated Flood 1996 Saguenay, QC 1.7 due to their slow, insidious development. However, as shown in Table 1, droughts were responsible for seven out Flood 1950 Winnipeg, MA 1.1 of the ten most costly natural hazards according to an Drought 1931-38 Prairies 1.0 analysis of natural hazards in Canada by Etkin et al (2005) (see Table 1). Drought 1989 Prairies 1.0

Based on the significance of drought and the extensive Hailstorm 1991 Calgary, AB < 1.0 impacts of the 1999-2005 drought, the concept of a 1 Cost is expressed in Billons 1999$.

1 University of Manitoba, Winnipeg 4 Prairie Farm Rehabilitation Administration, Winnipeg

2 Prairie Farm Rehabilitation 5 Manitoba Water Stewardship, Administration, Saskatoon Winnipeg

3 Saskatchewan Research Council, Saskatoon 6 Manitoba Hydro, Winnipeg

-97- CMOS Bulletin SCMO Vol.36, No.3 2. Impacts of and lessons from the 1999-2005 drought water and for drought information. Analysts must balance Unlike many previous droughts when only one or two the concerns and drought criteria from different sectors as economic sectors experienced impacts, the 1999-2005 they provide their recommendations to senior government drought affected agriculture, recreation, tourism, health, officials. As a result they need a better understanding of the hydro-electricity production, and forestry. According to a adequacy of various indices in different regions and preliminary analysis of the 2001 and 2002 drought years seasons. over Canada (Wheaton et al., 2005; Wheaton et al., 2008) the Gross Domestic Product was reduced by about $5.8 The Government of Canada’s Drought Watch, the North billion for 2001 and 2002. Other impacts included: American Drought Monitor and the Prairie Farm Rehabilitation Administration’s (PFRA) National Agroclimate # Employment losses that exceeded 41,000 jobs during the Information Service are examples of monitoring programs two-year period. that can provide a basis for informed decisions. However, users report that the coarse time and space resolution of # A negative or zero net farm income for several provinces the drought monitor limits its utility for some local for the first time in 25 years (Statistics Canada, 2003) with applications. Even widely used drought indices require agricultural production over Canada dropping an estimated further assessment. For example, the Palmer Drought Index $3.6 billion in 2001/2002. Farm input suppliers were hit by which is used extensively in temperate environments may lower demand for their products, while food processors not work as well at more northern latitudes or in the cold experienced shortages of raw material. season. Other indices, such as the Standardized Precipitation Index, may need improvement for more arid # Previously reliable water supplies such as streams, regions. wetlands, dugouts, reservoirs, and groundwater were placed under stress and often failed. At the provincial level, the drought monitoring task is made easier when a province has a drought plan in place that # In British Columbia and Manitoba, hydroelectric contains thresholds for identifying the onset of drought. generation was severely impacted, necessitating the Some provinces, such as Alberta, British Columbia and operaton of backup thermal resources and purchases of Ontario, have objective criteria for specifying drought power from neighbouring jurisdictions. conditions, although these criteria may not account for all of the complexities of drought processes. Developing these # In 2002, the incidence of forest fires in Alberta increased criteria is a challenge because of a lack of funds for to five times the ten-year average. research and monitoring and some inertia in agreeing to adopt a single unified objective strategy of using drought # At least 32 incidents of massive dust storms with criteria for programmatic decisions. associated traffic accidents including several fatalities were reported in Saskatchewan between April and September To meet the specific needs of individual provinces, each 2001. prairie province has taken steps to augment the services available from the federal government. For example, Alberta 3. User needs and the application of DRI information Agriculture maintains an operational modeling effort to There is a clear need to augment our understanding of estimate drought conditions which quantifies soil moisture drought events and to make this information available to in time and space. The Alberta program provides good those who can plan effective responses to drought. DRI has quality-monitoring information and advisory services, promoted user involvement and interaction with its science although concerns exist about the representativeness of team at each of its three annual workshops. From these certain indices and the consequences of using a variety of interactions and deliberations of the recently formed DRI interpolation schemes in cases where spatial variability is Partners Advisory Committee (PAC) a profile of user large. In addition, there are limitations in the availability of requirements for drought information has been developed. accurate comprehensive data related to evapotranspiration, The key requirements are summarized under four general soil moisture and precipitation, which all are needed to drive need statements. and/or validate their drought monitoring models. Wintertime information, including snow pack and snow melt information Need #1: Users need to know when they are in drought also have been identified as important for monitoring soil conditions. moisture and ground water recharge. Other provinces approach drought monitoring and response in support of the Drought impacts are often non-structural but can affect agricultural sector differently, but have similar information areas of thousands of square kilometres. Knowing when a requirements. For example, the Canadian Wheat Board has region is in a drought is fundamental for planning and played a lead role in working with the provinces to develop implementing adaptation measures and compensation a real-time meso-scale network in western Canada. programs. To provide more effective assessments of a drought it is essential to know an area’s sensitivity to dry Information needed to support real-time water resource conditions throughout the annual cycle. As noted earlier, management decisions at the provincial or utility level relies different sectors have different seasonal requirements for largely on a knowledge of the surface and subsurface water

CMOS Buletin SCMO Vol.36, No.3 -98- stores as well as streamflow and soil moisture conditions. potential and implications for droughts as well as reservoir Many of the surface hydrology variables necessary to water supply projections and river flow and level forecasts support agriculture are also needed for monitoring water during times of drought. Forecasts are often prepared based availability and hence the collection and use of information on statistically derived information on future weather needs to be closely coordinated across sectors. For conditions. There is a concern that such statistics may not example, coordination of networks in Manitoba involves be representative for some climatological situations and Manitoba Water Stewardship, Environment Canada, should be modified for an El Niño condition or for the onset Manitoba Agriculture, Manitoba Hydro and other Manitoba of a drought. To meet these needs DRI is exploring the partner agencies working together to operate and improve effects of sea surface temperature and soil moisture climatic, hydrometric and soil moisture networks for use in assimilation on seasonal drought prediction. DRI also is monitoring and forecasting provincial water supplies. exploring the various tools (e.g. satellite data products, forecast evaluation procedures) that might be useful in DRI contributes to these efforts through its assessment of assessing seasonal climate prediction products for western various drought indices and its research on the Canada. These activities are described in more detail in characterization of drought. Several projects use provincial Stewart et al. (2008) and by reports and presentations data sets, thus allowing DRI to assist in clarifying the available through the DRI website: www.drinetwork.ca. potential uses, benefits and limitations of these regional data systems and their derived products. Need #3: Users need to know the probability of drought events based on an understanding of physical processes Need #2: Users need reliable forecasts of precipitation and other critical variables that relate to the initiation, For planning purposes it is important to know the recurrence intensification, continuation and termination of droughts. interval of major drought events. Water managers responsible for dam and reservoir construction need access Both the agriculture community and the water resources to studies similar to those done by DRI for other historical community need forecasts indicating when droughts will droughts that have affected regional water supplies and begin and end. For example, during periods of extended hydropower production. Many planners use a single year or system-wide drought, Manitoba Hydro’s hydropower multiple years from the dustbowl of the 1930s as the generation may be severely impacted. Consequently, benchmark worst event against which they evaluate their reliable forecasts of both the onset and termination of major plans. For example, Manitoba Hydro uses a year in the system-wide drought events are critical for Manitoba Hydro 1930s decade as the “drought of record” to plan the system to ensure that the operation of its integrated system is so that sufficient future resources are developed, ensuring optimized. Seasonal forecast skill is limited by the inherent they can supply the firm demand even under the expected variability of Canada’s climate and by the challenges faced worst drought conditions. However, the 1930s are difficult by dynamic prediction models when they are used for to address with detailed physical studies and seasonal forecasting. Some opportunities to develop comprehensive documentation because the data networks prediction skill may be inferred by studies of teleconnection were much less developed than they are today and the patterns such as the work done by Shabbar (1997) and number of reanalysis and model simulations available for others. DRI studies related to atmospheric circulation the 1930s are limited. Manitoba Hydro is also assessing patterns during drought periods are described by Bonsal paleoclimate records (e.g. tree rings, lake sediments) within (2008) in this issue. their watershed as another source of information to understand more about extreme droughts and climate At the provincial level the needs for forecast variables are variability in past climates (see Sauchyn et al., 2002). linked to the criteria used to monitor droughts. For example, Manitoba Hydro would benefit from a long-term broad scale provinces that rely on soil moisture to monitor drought analysis of drought response using threshold criteria that would like accurate soil moisture forecasts to assess are compatible with their operations to see how climate possible future drought conditions as well as improved in- variability affects “hydroelectric” droughts over their region. season forecasts of factors such as precipitation and evaporation that drive changes in soil moisture. Many users are concerned about potential increases in the Increasingly, seasonal forecast groups are beginning to frequency and intensity of droughts in the future. Some advocate the use of ensemble forecasts, although some climate models indicate that the frequency of extreme users have questioned the utility of such forecasts. The events such as floods and droughts will increase with hesitancy by some to use ensemble forecasts seems to increasing concentrations of atmospheric carbon dioxide. stem from a lack of information and confidence about the Estimations of the probability of multi-year droughts are type and accuracy of information they provide and how particularly needed as these events have some of the these forecasts can be applied to their particular needs. largest negative impacts. This is very important for provincial water managers who need to plan and allocate Forecasts are necessary to guide decisions on how to water in a sustainable fashion in order to minimize the risk operate dams and address issues of water availability for of water shortages for towns, industries, irrigation and downstream uses. A number of provincial water individuals. Many believe that uncertainties associated with management agencies issue outlooks that review the the implications of long-term climate change for drought

-99- CMOS Bulletin SCMO Vol.36, No.3 remain sufficiently large that the best course of action lies The impacts of drought are often regional in nature. In order in preparing for potentially increasing climate variability. For to target mitigation strategies that are most appropriate it is example, PFRA is working on a planned approach for important to assess the area’s vulnerability to drought. The agriculture to develop tools and policies to reduce drought risk associated with a drought is dependent on the threat of impacts; to support and develop current activities that work, the event (representing the climate and any other factors and to develop and implement new activities to adapt to that could affect the water supply) and the vulnerability to drought. It is recognized that federal activities would be drought (representing the economic and engineered most effective if they are linked to the provinces by infrastructure in a region). As a first approximation areas sustained institutional partnerships. It also is important to which rely heavily on water would be more susceptible to ensure that the key factors associated with these potential drought although Venema (2007) has shown that changes in the characteristics of drought are carefully vulnerability and resilience contain a strong learning monitored so any shifts in climate can be detected and component because many farmers in drought-prone areas users can receive early notification of such changes. have learned to cope with drought better than those in areas where drought occurs irregularly. In some ways the DRI multi-year drought seemed to be unique in that it persisted through a number of different sea- For historical and jurisdictional reasons, drought surface temperature and teleconnection patterns: some of management plans are frequently developed for provinces which have occurred on other occasions without producing but they can be made even more specific for a region or a such significant droughts. While DRI has not been funded city. While governments can put in place an economic to look at comparisons of the 1999-2005 drought with other safety net, the decisions about drought response often are droughts, some work is being carried out (Bonsal, 2008). An most effective when they are agreed to by those who will be understanding of the processes that cause the droughts implementing the plans. based on DRI analysis could enable more insightful assessments of extremes in future studies because it will During droughts, many provinces activate provincial inter- provide a stronger basis for exploring the frequency of the departmental drought committees which coordinate causative factors that lead to drought than a purely provincial responses to droughts. They also develop statistical analysis of historical and projected values of newsletters and other communication tools in order to temperature and precipitation. provide information on the drought and options for water conservation. Most provinces rely on federal information Need #4: Users need guidance on how to use physical systems as well as data from their own networks, such as understanding and insights to reduce vulnerability to provincial groundwater observation networks, to support drought and mitigate drought impacts. their decisions regarding drought response. These committees encourage drought responses tailored to There are needs for inputs to policy and programmatic provincial needs. For example, the Manitoba Water responses to drought at all levels of government including Services Board helps farmers acquire the necessary water the municipal level. For the agricultural sector, Canada supplies, such as pumps, when a drought is developing, needs to take a proactive stance on agricultural drought to ensure it maintains its competitive advantage in world food The development of a risk-based approach to drought markets through its natural climate advantage and its management can integrate management considerations on historic government support to this sector. This competitive all scales. These approaches should include internal risk advantage can be best achieved by long-term research, assessment and sensitivity analyses of more severe investments, and policies. Historically, governments have droughts (Botterill and Fisher, 2003; NDPC, 2000). As responded to agricultural droughts by supporting a range of described by Wilhite et al. (2005), countries such as South on-farm actions such as water supply development, water Africa, Australia and the United States are developing these conservation, feed support for livestock and education approaches for drought management. Risk assessment programs to name a few. While these programs are should be supported by reliable drought prediction beneficial for individual farmers the benefits of a broader capabilities and monitoring systems. However, these strategic approach for the sector should also be developed. approaches must ensure that the needs and concerns of all To this end, federal-provincial committees are developing a levels of government are addressed. DRI can provide framework for the next generation of agricultural policy research support at each of these jurisdictional levels. For framework, known as Growing Forward, as well as a example, DRI hydrological process research on sublimation national drought strategy for the medium- to long-term. and blowing snow can provide practical advice on stubble Drought management and adaptation for longer-term management for optimizing snowpack retention through the climate change will continue to receive attention in federal- winter months. On the larger scale, innovations developed provincial discussions and the research work of DRI will through DRI for using satellite and in-situ data should assist have continued relevance to, and benefit for, understanding in developing a stronger scientific basis for risk-based drought, contributing to policy and sustaining Canada’s approaches. economic development.

CMOS Buletin SCMO Vol.36, No.3 -100- 4. The Role of the Partners Advisory Committee climates in a national and, possibly, an international context. The DRI Network seeks to connect its research to practical In terms of the societal impacts of drought, questions such applications wherever appropriate. Three opportunity areas as: “how are organizations affected by drought?”, “what is of focus have been identified: data products, analytical tools the nature of these impacts?”, and “how can drought and models; improved forecasts, and syntheses that can be impacts be alleviated?” could be addressed. DRI is used for policy development. In order to help focus the committed to continuing its dialogue with users to obtain applications aspects of DRI, a small advisory committee inputs and advice as it develops its second phase proposal. made up of some of the DRI Partners has been established to provide advice to the DRI Board of Directors (BoD). With 6. Acknowledgements: this level of reporting it is believed that the advice will The authors wish to thank the Canadian Foundation for influence DRI research at its most fundamental levels. Climate and Atmospheric Research (CFCAS) for their support to DRI and for enabling the research-user dialogue The Terms of reference for this committee are as follows: that is provided by the DRI framework. In addition, each of the authors thank their home institutions for their support 1) To provide advice and assessments to the BoD while contributing to this paper, while the lead author thanks on the status of DRI interactions with stakeholders; DRI and CFCAS for their support. We also wish to thank the various DRI partners who generously shared their views on 2) To identify linkages and propose working groups drought research needs. In addition, thanks go to Rick and recommend actions that could be taken to Raddatz for his editorial comments. develop specific proposals; and 7. References: 3) To develop an outreach strategy to ensure DRI results are used effectively. Bonsal, B., 2008: Droughts in Canada: an overview. CMOS Bulletin SCMO (in this issue). Membership on the committee includes representatives from federal agencies and from water management and Botterill, L., M. Fisher, eds., 2003: Beyond Drought: People, agricultural agencies in western Canadian provinces. One Policy, and Perspectives. Melbourne: CSIRO Publishing. potential area of activity involves preparing for droughts Etkin, D., Haque, E., Bellisario, L., & Burton, I., 2004: An using scenarios and simulation exercises. The purpose of Assessment of Natural Hazards and Disasters in Canada: A these activities is to identify gaps in current capacity to Report for Decision-Makers and Practitioners (PDF Version). respond to drought and to better understand potential Prepared for The Canadian Natural Hazards Assessment Project. challenges arising from demographic and economic trends. ISBN 0-9735436-0-4. PFRA’s Climate Decisions Support and Adaptation Unit is working with DRI to explore how to develop prototypes http://www.crhnet.ca/reports/Hazards_Assessment_Summ related to this activity. ary_eng.pdf

Maybank, J., B. Bonsal, K. Jones, R. Lawford, E.G. O'Brien, E. 5. Summary and Next Steps Ripley, and E. Wheaton, 1995: Drought as a natural disaster. Part This review of the user needs has identified a number of of a special volume for the International Decade on Natural areas where more research is needed. In particular, users Disaster Reduction, Atmosphere-Ocean, 33(2), 195–222. need to know: National Drought Policy Commission (NDPC), 2000: “Preparing for 1) information for determining when a region is in Drought in the 21st Century.” Washington, D.C. drought conditions, Sauchyn, David J., Antoine Beriault, and Jennifer Stroich. 2002. 2) reliable forecasts of drought and associated A Paleoclimatic context for the drought of 1999-2001 in the climate variables such as precipitation and other northern Great Plains. Geographical Journal : 1-18. drought variables. Shabbar, A., Bonsal, B. and Khandekar, M. 1997. Canadian precipitation patterns associated with the Southern Oscillation. 3) the probability of causative factors for droughts Journal of Climate, 10: 3016-3027. of different temporal or spatial extent. Statistics Canada, 2003: Net Farm Income, Ottawa, ON, 4) ways to apply drought research results to Catalogue No. 21-010-WIE 2(1). reducing vulnerability to drought and to mitigating its impacts. Stewart, R., J. Pomeroy, and R. Lawford, 2008: A Drought Research Initiative for the Canadian Prairies. CMOS Bulletin SCMO DRI has made significant progress in addressing the first (in this issue). two needs and has made some contributions to the third Venema, H., 2007: Adaptation as Resilience Building: A policy and fourth issues. DRI is seeking opportunities for a follow- study of climate change vulnerability and adaptation on the on study that will provide opportunities for addressing Canadian prairies and Implications for DRI. 2nd Annual DRI droughts and other extremes in both past and future Workshop, Winnipeg, Manitoba, January 11, 2007.

-101- CMOS Bulletin SCMO Vol.36, No.3 Wilhite, D.A., L. Botterill, and K. Monnik, 2005: “National Drought Appendix A Policy: Lessons Learned From Australia, South Africa, and the United States.” Drought and Water Crises: Science, Technology, Members of the DRI Partner Advisory Committee and Management Issues, Edited D. A. Wilhite, Taylor and Francis Group. Ryan Cossitt (Saskatchewan Agriculture and Food) Wheaton, E., V. Wittrock, S. Kulshreshtha, G. Koshida, C. Grant, Bill Girling (Manitoba Hydro) A. Chipanshi, and B.R. Bonsal, 2005: Lessons Learned from the Canadian Drought Years of 2001 and 2002: Synthesis Report, Irene Hanuta (Agriculture and Agri-Food Canada) Agriculture and Agri-Food Canada, Saskatchewan Research Council Publication No. 11602-46E03, Saskatoon, SK. Harvey Hill, Chair (Agriculture and Agri-Food Canada)

http://www.agr.gc.ca/pfra/drought/info/11602-46E03.pdf Ray Keller (Alberta Environment)

Wheaton, E., S. Kulshreshtha, V. Wittrock and G. Koshida, 2008: Rick Lawford (ex-officio) Dry times: hard lessons from the Canadian drought of 2001 to 2002. The Canadian Geographer (Accepted for Publication). Andrew Nadler (Manitoba Agriculture, Food and Rural Initiatives)

Bart Oegema (Saskatchewan Watershed Authority)

Alf Warkentin (Manitoba Water Stewardship)

Elaine Wheaton (Saskatchewan Research Council)

Ralph Wright (Alberta Agriculture and Food)

Kenneth Korporal (Secretariat for the Canadian Group on Earth Observations, Environment Canada)

Description de la page couverture Trois articles présentés dans ce numéro traitent des sécheresses au Canada tout en favorisant la continuité des Ce numéro: Les sécheresses prolongées sur grande objectifs de l’équipe du réseau DRI. Ces articles donnent superficie sont parmi les catastrophes naturelles les plus une vue d’ensemble sur les sécheresses au Canada et met coûteuses au Canada. Elles ont des répercussions l’accent sur celles des Prairies canadiennes (page 79). De profondes sur tout un éventail de secteurs, notamment sur plus, une description des aspects du réseau DRI suit (page l’agriculture, les forêts, l’industrie, les municipalités, les 87). En conclusion, on évalue les attentes des utilisateurs loisirs, la santé, la société et les écosystèmes aquatiques. de la communauté du réseau DRI, incluant leurs besoins Bien que la plupart des régions du Canada doivent faire particuliers (page 97). face à la sécheresse, les Prairies canadiennes sont les plus vulnérables avec au moins 40 sécheresses qui sont Page Couverture: La photo en page couverture a été prise survenues depuis deux siècles. Mais rarement cette près de Consul, Saskatchewan, en avril 2002. Elle a été sécheresse a-t-elle suscité autant d’inquiétudes que lors de prise par Ted Banks, Agriculture et Agroalimentaire l’épisode 1999-2005 qui a abouti à la pire sécheresse qui Canada. soit survenue au Canada depuis un siècle, touchant plus particulièrement les Prairies canadiennes. Pour traiter de CMOS Executive Office / Bureau de la SCMO telles sécheresses extrêmes, un réseau de recherche du nom de DRI (Drought Research Initiative – Initiative de Dr. Ian Rutherford recherche sur la sécheresse) a été formé. Le centre Executive Director - Directeur exécutif d’intérêt du réseau DRI consiste à mieux comprendre la Tel/Tél.: 613-990-0300 façon dont les sécheresses surviennent, évoluent et E-mail/Courriel: [email protected] prennent fin afin de comprendre leur structure interne et de contribuer à une meilleure prédiction de tels événements. Dr. Richard Asselin Afin de réaliser cet objectif, on considère la sécheresse Director of / Directeur des Publications sous différents points de vue en tenant compte de Tel/Tél.: 613-991-0151 l’atmosphère, du sol incluant les processus de la végétation E-mail/Courriel: [email protected] et du sous-sol. De plus, la communauté de recherche du réseau DRI travaille étroitement avec plusieurs partenaires Ms. Qing Liao concernés par la sécheresse et qui ont tous intérêt à faire Office Manager - Chef de bureau face à de tels événements dans le futur. Tel/Tél.: 613-991-4494 E-mail/Courriel: [email protected]

CMOS Buletin SCMO Vol.36, No.3 -102- Improved seawater thermodynamics: How should the proposed change in salinity be implemented?

by SCOR/IAPSO Working Group 127

February 2008

Background The SCOR/IAPSO Working Group 127 on the “Equation of available estimate of the Absolute Salinity of both State and Thermodynamics of Seawater” is charged with Reference Seawater and the Standard Seawater that was providing improved algorithms and descriptions of the used in the measurements of the physical properties. From thermodynamic properties of seawater. The working group a practical point of view, the value of SR can be related to has made significant progress on many of its goals, and it the Practical Salinity S by is now time to seek the advice of the oceanographic -1 community regarding the best practical ways of adopting SR = (35.165 04 / 35) g kg X S. these developments into oceanographic practice. The Working Group has met twice to date, once in Warnemünde Reference Seawater that has been normalized to a in 2006, then in Reggio Calabria in 2007. Our next meeting Practical Salinity of 35 has a Reference-Composition -1 is in Berlin in September 2008. Salinity of exactly SR = 35.165 04 g kg .

The working group will soon provide the most accurate The new independent salinity variable SR is intended to be algorithms to date for the thermodynamic properties of used as the concentration variable for future thermodynamic seawater (such as density, entropy, enthalpy, specific heat functions of seawater, as an SI-based extension of Practical capacity, etc). In order to achieve such accuracy it became Salinity, as a reference for natural seawater composition evident that a salinity variable is required that more anomalies, as the currently best estimate for Absolute accurately represents absolute salinity than does the Salinity of IAPSO Standard Seawater, and as a theoretical conductivity-based Practical Salinity. Spatial variations in model for the electrolyte mixture “seawater”. the composition of seawater upsets the relationship between Practical Salinity S (which is a function of As described in this abstract, for seawater of standard conductivity, temperature and pressure) and Absolute composition we have been able to relate the Absolute

Salinity SA (defined as the mass of dissolved material per Salinity to the Practical Salinity; for example, at a Practical mass of seawater solution). If the thermodynamic properties Salinity of 35, seawater of Reference Composition has an of seawater are to be written in terms of just one type of Absolute Salinity of 35.165 04 g kg–1. We expect shortly to salinity, then they are much closer to being functions of (SA, be able to recommend an algorithm that accounts for the t, p) than being functions of (S, t, p). Moreover, Absolute variation of seawater composition from the standard Salinity is a conservative property (that is conserved when composition. That is, we soon expect to be able to turbulent mixing occurs) whereas Practical Salinity is not recommend an algorithm SA = SA(SR, , , ,) where the extra conservative. arguments will be either measured parameters (such as total alkalinity, silicate and nitrate) or more simply the spatial Absolute salinity for seawater of Reference locations longitude, latitude and pressure. Millero and Composition Kremling (1976), Millero (2000) and Millero et al (2008b) In order to progress toward evaluating Absolute Salinity our are precursor papers to such an algorithm. first task was to define the relative concentrations of the constituents of Standard Seawater. This we have done, Advantages of Absolute Salinity over Practical Salinity and this work is published in Millero et al (2008a). The Absolute Salinity has the following advantages over abstract of this paper is as follows. Practical Salinity for oceanographic use.

Fundamental determinations of the physical properties of 1. The definition of Practical Salinity S on the PSS-78 seawater have previously been made for Atlantic surface scale is separate from the system of SI units. waters, referred to as “Standard Seawater”. In this paper a Absolute Salinity can be expressed in the unit (g Reference Composition consisting of the major components kg–1). Adopting this SI unit for salinity would of Atlantic surface seawater is determined using these terminate the ongoing controversies in the earlier analytical measurements. The stoichiometry of sea oceanographic literature about the use of “psu” or salt introduced here is thus based on the most accurate “pss” and make research papers more readable to prior determination of the composition, adjusted to achieve the outside scientific community and consistent charge balance and making use of the 2005 atomic weights. with SI. Reference Seawater is defined as any seawater that has the Reference Composition and a new Reference-

Composition Salinity SR is defined to provide the best

-103- CMOS Bulletin SCMO Vol.36, No.3 2. The freshwater mass fraction of seawater is not (1 Salinity to our “salinity” variable, which to date has been 1 – 0.001 S). Rather, it is (1– 0.001 SA /(g kg )), Practical Salinity. For example, if we were intent on where SA is the Absolute Salinity, defined as the interpreting the salinity of an ocean model as Practical mass fraction of dissolved material in seawater. Salinity, then the salt conservation equation should contain –1 The values of SA /(g kg ) and S are known to differ a non-conservative source term to take account of the by about 0.5%. There seems to be no good reason spatial variations in the composition of seawater. for continuing to ignore this known difference, e.g., in ocean models. Here we summarize the reasons why Absolute Salinity is the preferred salinity variable for oceanographic research. 3. PSS-78 is limited to the salinity range 2 to 42. For a smooth crossover on one side to pure water, and # It will be preferred by journals since it is an SI unit. on the other side to concentrated brines up to saturation, as e.g. encountered in sea ice at very # It is the natural salinity variable for ocean models since low temperatures, salinities beyond these limits they assume that their salinity variable is conservative, need to be defined. While this poses a challenge hence it should be used to initialize ocean models at all

for S, it is not an issue for SA. depths.

4. The theoretical Debye-Hückel limiting laws of # It is the natural variable to use in inverse models, budget seawater behaviour at low salinities, used for studies and on salinity-temperature diagrams because its example in the determination of the Gibbs function conservative nature justifies turbulent mixing occurring of seawater, can only be computed from a chemical along straight lines on such a diagram.

composition model, which is available for SR but not for S. # The freshwater fraction and the meridional freshwater flux follow naturally when using Absolute Salinity but not when 5. For artificial seawater of Reference Composition, using Practical Salinity.

SR has a fixed relation to Chlorinity, independent of conductivity, salinity, temperature, or pressure. # By using Absolute Salinity in the algorithm for the equation of state, the effects of the spatial variations of 6. The next largest improvement in the equation of seawater composition are accounted for, while if Practical state of seawater will come from incorporating Salinity is used in such a call to the equation of state, a variations in the composition of seawater, that is, density error is incurred. from calling the equation of state with Absolute Salinity rather than with Reference Salinity. The # It is the common salinity variable used in engineering, determination of Absolute Salinity is facilitated by natural and geosciences outside oceanography, where the introduction of the Reference Composition and Practical Salinity is often unknown or misconstrued. Reference Salinity. # It is applicable to low concentrations in brackish lagoons

7. Absolute Salinity SA is a conservative variable, and river mouths, to high concentrations in freezing or whereas, in the presence of compositional desiccating brines, as well as at higher temperatures in variations, Practical Salinity S (which is essentially desalination plants, whereas Practical Salinity is defined determined by conductivity alone) is not a only in the range 2

CMOS Bulletin SCMO Vol.36, No.3 -104- SA = SR + δSA Given this, is it worthwhile changing the present archiving practice in favour of a variable (Reference Salinity) that is where Reference salinity SR is simply proportional to still not the final salinity that we will use (Absolute Salinity)? Practical Salinity S as described in Millero et al (2008), namely by Any choice of action inherently involves compromises, and the best course of action is not obvious to the Working –1 SR = (35.165 04 / 35) g kg X S , Group. As a way of focussing the discussion we outline two possible routes for adopting the advantages of Absolute and δSA is the difference between Absolute and Reference Salinity, labelled Option 1 and Option 2. Salinities. δSA will be available as a look up table as a function of latitude, longitude and pressure and also as an Option 1 alternative linear relationship of nutrient and silicate # Change from reporting Practical Salinity to reporting concentrations, or for example, as a Calcium excess Reference Salinity to national and estimate from the river discharge into estuaries. We expect international data bases. This implies that the data bases to have algorithms available before the end of 2008. store Practical Salinity from the old cruises and store Reference Salinity from new cruises (from say 1st January How to adopt Absolute Salinity? 2010). Having made the case that Absolute Salinity possesses many advantages over Practical Salinity, how should # Provide software [for example, of the form SA(SR,x, y, p)] present oceanographic practice adapt to incorporate these to produce the best available estimate of Absolute Salinity advantages? from Reference Salinity (using additional information on position or water properties). The obvious thing to do would be to decide on a date on which the whole community ceases to use Practical Salinity # Have all the thermodynamic software in the form and switches to using Absolute Salinity. However the ρ (SA,t,p). algorithm to convert Reference Salinity to Absolute Salinity is less mature and will probably remain a “work in progress” Discussion of Option 1 for several years. Moreover, data that are stored in archives The main advantage of Option 1 is that the community should have a very close connection to a measurement (like eventually ceases to use the non-SI unit Practical Salinity, temperature or conductivity) rather than being the result of and instead uses the two SI salinity measures, Reference an algorithm that is likely to change with time. Hence one Salinity and Absolute Salinity. cannot really imagine storing Absolute Salinity in data bases. Rather, the closest thing to do in this vein is to store A drawback of Option 1 is that there will be cases of Reference Salinity. contamination of the data bases where cruise salinity is labelled and stored as Reference salinity whereas in fact it Storing Reference Salinity in data centres would have the is Practical Salinity data, and vice versa. This kind of error advantage that it is an SI unit. However before the equation presently contaminates the temperature, oxygen and of state (or other thermodynamic quantities) can be pressure/depth data bases. evaluated using the new software, the Reference Salinity data needs to be converted to Absolute Salinity using the Since both S and SR are simply measures of conductivity, most up-to-date version of this software. Moreover, the and since they are simply proportional to each other, will it community cannot completely abandon Practical Salinity be seen that we are taking a course of action that has since it will remain as the salinity variable in the archives for potential for confusion for only academic benefit? cruises undertaken before the change-over date. By changing the salinity variable that is reported from cruises Recall that scientific work and papers are mostly done with to data bases from Practical Salinity to Reference Salinity potential temperature θ rather than in situ temperature t so the possibility of contamination of the data archives arises the first thing that one usually does with the S, t, p data from as salinity of one type is incorrectly labelled and stored as a data centre is to form θ. Similarly, scientific work and the other type of salinity. papers should be done with Absolute Salinity rather than Reference Salinity so the first thing that one needs to do

In the long run, as with many other historical non-SI units under Option 1 with the ,SR t, p data from a cruise or from a like torr, cal or dyn, it would seem to be an advantage to data centre is to form not only θ but also SA . This analogy use only Reference Salinity and abandon the use of with what we already do with storing the measured variable Practical Salinity completely. If Reference Salinity were the t but using the derived variable θ is very close. salinity variable to be used in all of the revised thermodynamic algorithms, the argument for “biting the Under Option 1 we cannot imagine that the community can bullet” and abandoning Practical Salinity as much as altogether forget about Practical Salinity however, as the possible would seem to be the correct path. But it is data from older cruises (e.g. all of WOCE) are stored in data Absolute Salinity that we seek, and Reference Salinity is centres in terms of Practical Salinity. These data will need only part way towards the evaluation of Absolute Salinity. converting first to Reference Salinity and then to Absolute

-105- CMOS Bulletin SCMO Vol.36, No.3 Salinity before the thermodynamic routines such as will affect the results and the publications arising out of potential temperature, density, potential enthalpy etc, can those who make this error, but this error will not be called by oceanographic researchers. contaminate an archived data set.

There will be some instances when the new software is Option 2 does not require manufacturers (such a Seabird called with the salinity data being S and in those instances and the Standard Seawater Service) to change what they an error will be made. This type of error is an undesirable presently do. Rather, Option 2 puts the responsibility for the consequence of both Options 1 and 2. changes in the hands of practicing research oceanographers. Option 1 requires manufacturers (such as Seabird) to change what they presently do. The instruments will need Request for your input to output their salinity in terms of Reference Salinity. Also The above two options are just two of many options; please the ampoules of standard seawater will need to quote their do not feel constrained in your comments to these options. salinity in terms of Reference Salinity. The transition date of We seek input from the oceanographic community on how say 1st January 2010 has to be handled very carefully in to gain the advantages of adopting Absolute Salinity in our these respects. Further, anyone wanting to make use of oceanographic research work. The key issue seems to older ampoules will have to be aware of the transition and revolve around which type of salinity is required to be how to deal with it. reported to and archived by oceanographic data centres. We encourage frank responses. Each response will be Option 2 thoughtfully considered by the Working Group. Please email # Continue to report Practical Salinity S from cruises and to your comments to [email protected] with the have only Practical Salinity S words “Comment for WG127 on how to adopt Absolute stored at national and international data centres. Salinity” as the message title.

# Provide software [for example, of the formSA(S,x, y, p)] to References produce the best available estimate of Absolute Salinity Feistel, R., 2008: A Gibbs Function for Seawater Thermodynamics –1 from Practical Salinity (using additional information on for !6 °C to 80 °C and Salinity up to 120 g kg . Submitted to Deep-Sea Research position or water properties. I, November 2007. Millero, F. J., 2000: Effect of changes in the composition of ρ S t p Have all the thermodynamic software in the form ( A, , ). seawater on the density-salinity relationship. Deep-Sea Research, 47, 1583-1590. Discussion of Option 2 By reporting only S in data bases we would expect to Millero, F. J., R. Feistel, D. G. Wright and T. J. McDougall, 2008a: greatly reduce the possibility of salinity data being The composition of Standard Seawater and the definition of the mislabeled in data bases. Reference-Composition Salinity Scale. Deep-Sea Research I, 55, 50-72.

Since both S and SR are simply measures of conductivity, option 2 is consistent with the argument that there is little Millero, F. J. and K. Kremling, 1976: The densities of Baltic Sea Waters. Deep-Sea Research, 23, 1129-1138. value in replacing one measure of conductivity (namely Practical Salinity) with another (namely Reference Salinity) Millero, F. J., J. Waters, R. Woosley, F. Huang and M. Chanson, in data bases. Rather, under Option 2 data centres store S 2008b: The effect of composition on the density of Indian Ocean and S alone. waters. Deep-Sea Research I, in press.

As mentioned above, scientific work and papers are mostly Membership of SCOR/IAPSO Working Group 127 on done with potential temperature θ rather than in situ “Thermodynamics and the Equation of State of Seawater” temperature t so the first thing that one usually does with the S, t, p data from a data centre is to form θ . Similarly, • Trevor J. McDougall, Chair, E-mail: [email protected] scientific work and papers will be mostly done with Absolute • Chen-Tung Arthur Chen, E-mail: [email protected] • Rainer Feistel, E-mail: [email protected] Salinity rather than Practical Salinity so the first thing that • Valentina N. Gramm-Osipova, E-mail: [email protected] one needs to do under Option 2 with the S, t, p data from • David R. Jackett, E-mail: [email protected] a cruise or from a data centre is to form not only θ but also • Brian A. King, E-mail: [email protected] SA . This analogy with what we already do with storing the • Giles M. Marion, E-mail: [email protected] measured variable t but using the derived variable θ • Frank J. Millero, E-mail: [email protected] suggests that storing S but using SA will not cause • Petra Spitzer, E-mail: [email protected] oceanographers any serious difficulties. • Dan Wright, E-mail: [email protected] Associate Member There will be some instances when the new software is • Peter Tremaine, E-mail: [email protected] called with the salinity data being S and in those instances an error will be made. This type of error is an undesirable Source: Canadian Ocean Science Newsletter, Vol.35, March consequence of both Options 1 and 2. However this error 2008. Reproduced here with the authorization of the Editor.

CMOS Bulletin SCMO Vol.36, No.3 -106- CMOS BUSINESS / AFFAIRES de la SCMO

A-O Abstracts Preview particulièrement marqué dans la mer d’Okhotsk et dans l’ouest de la mer de Béring, qui sont des régions sources possibles de production d’eaux intermédiaires du Pacifique Avant Première des résumés de A-O Nord glaciaires. Dans ces régions, une augmentation du rejet de saumure causé par une augmentation marquée des The following abstracts will soon be published in your glaces de mer, l’excès de l’évaporation sur les précipitations ATMOSPHERE-OCEAN publication. et une réduction des débits fluviaux contribuent à l’accroissement de la salinité en surface. En résumé, la Les résumés suivants paraîtront sous peu dans votre réduction de l’apport d’eau douce dans le Pacifique Nord revue ATMOSPHERE-OCEAN. septentrional est la principale raison de l’accroissement de la production et la sortie d’eaux intermédiaires du Pacifique Nord glaciaires. Glacial Ocean Circulation and Property Changes in the North Pacific

SEONG-JOONG and YOUNG-GYU PARK Cloud type observations and trends in Canada 1953-2003

Abstract E. J. MILEWSKA The glacial properties and circulation changes in the North Pacific Ocean are investigated using a coupled ocean- atmosphere-sea-ice climate model. With glacial boundary Abstract conditions, an increase in potential density in the upper The monitoring of cloud amount and type in Canada is layers of the northern North Pacific makes the water column discussed in detail, including observing, archiving, data highly unstable and eventually results in the enhancement transmission procedures and practices, and automation. of North Pacific Intermediate Water (NPIW) production, consistent with proxy evidence. The NPIW outflow reaches There have been some major monitoring challenges since deeper layers than in the present ocean, but remains largely 1953. In 1977, the network-wide replacement of detailed confined to the North Pacific. The increase in potential cloud layer amounts and obscuring phenomena by broad density is predominantly caused by the increase in salinity sky conditions, based on summation amounts, imposed and, to a lesser extent, by decreases in temperature. The analysis of frequency of occurrence of mainly cloudy increase in surface salinity is especially high in the Sea of conditions rather than actual amounts. Partial automation Okhotsk and the western Bering Sea, which are possible with Automated Weather Observing Systems resulted in the source areas of glacial NPIW production. In these regions, cessation of observations of higher clouds and cloud types, an increase in brine release due to a marked increase in as well as the incompatibility of sky coverage with human sea ice, evaporation exceeding precipitation, and a observations at eight percent of stations. reduction in river discharge contribute to the increase in surface salinity. In short, reduction in freshwater input to the For every hourly report from eighty-four airport stations, northern North Pacific is the main reason for the increase in from 1953 to 2003, each layer is classified according to the production and outflow of glacial NPIW. cloud type and related standard base height into three levels of clouds, low, middle and high clouds. Trends in Résumé occurrence of summation amounts of mainly cloudy Nous étudions les changements dans les propriétés et la conditions at each of these three levels are computed on circulation glaciaire dans l’océan Pacifique Nord au moyen annual, seasonal, daytime and nighttime scales, together d’un modèle climatique couplé océan-atmosphère-glaces with annual trends in occurrence of selected convective and de mer. Avec des conditions glaciaires aux limites, une stratiform clouds. Based on annual anomalies averaged augmentation de la densité potentielle dans les couches over the country and provinces, no major network-wide supérieures du Pacifique Nord septentrional rend la colonne systematic discontinuities were noted; on average, on an d’eau très instable et entraîne éventuellement une annual basis over the entire network, slight decreasing accélération de la production d’eaux intermédiaires du trends are noted for summation amounts of mainly cloudy Pacifique Nord, qui s’accorde avec les preuves indirectes. conditions at low and middle levels, and increasing trends La sortie d’eaux intermédiaires du Pacifique Nord atteint at high levels. des couches plus profondes que dans l’océan actuel mais demeure largement confinée au Pacifique Nord. The increasing trend at high levels is indeed remarkable. L’accroissement de densité potentielle est principalement The rate of increase, especially rapid until 1974, has been causé par l’accroissement de la salinité et, dans une shown to be caused by a prominent increase in cirrus cloud moindre mesure, par des diminutions de température. reports. The link between this rise and the increase in air L’accroissement de la salinité en surface est traffic was established by others in the United States. This

-107- CMOS Bulletin SCMO Vol.36, No.3 link may also apply in Canada, which experienced a similar La tendance à la hausse dans les niveaux élevés est en fait expansion in aviation. remarquable. Le taux d’accroissement, particulièrement rapide jusqu’en 1974, a été expliqué par un accroissement Notably, the largest increase in high nighttime cloudiness important des observations de cirrus. D’autres chercheurs and decrease in low-middle cloudiness is evident in western aux États-Unis ont établi le lien entre cet accroissement et Canada, possibly contributing to the recently observed l’accroissement du trafic aérien. Ce lien peut aussi warming of daily minimum and maximum temperatures s’appliquer au Canada, où le secteur de l’aviation a connu there. une expansion semblable.

The occurrence of stratiform clouds at all levels exhibits Détail d’intérêt, la plus forte augmentation de nuages élevés significant decreasing trends across the country, except for la nuit et diminution de nuages bas ou moyens se produit southern Ontario. Clouds of intense convection show dans l’ouest du Canada, ce qui contribue peut-être à la pronounced decreasing trends in western Canada, while not hausse des températures journalières minimales et much change is evident elsewhere. Similar to cirrus, maximales récemment observée dans cette région. stratocumulus is notable as it shows strong positive trends everywhere in the country. On the other hand combined L’occurrence de nuages stratiformes à tous les niveaux stratus and stratus fractus clouds exhibit decreasing trends accuse une tendance à la baisse appréciable dans tout le except over British Columbia where the opposite occurs. pays sauf dans le sud de l’Ontario. Les nuages de The findings concerning stratocumulus, stratus and stratus convection intense affichent une tendance à la baisse fractus clouds in Canada are similar to the findings in the prononcée dans l’ouest du Canada alors qu’on ne constate United States. pas beaucoup de changement ailleurs. Comme pour le cirrus, le stratocumulus est remarquable par les fortes Résumé tendances positives qu’il affiche partout au pays. D’autre Nous discutons en détail de la surveillance de l’étendue et part, le stratus et le stratus fractus combinés affichent des du type des nuages au Canada, y compris l’observation, tendances à la baisse sauf en Colombie-Britannique où l’archivage, les procédures et pratiques de transmission des l’inverse se produit. Les conclusions tirées à propos du données et l’automatisation. stratocumulus, du stratus et du stratus fractus au Canada concordent avec les conclusions tirées aux États-Unis. La surveillance a posé d’importants défis depuis 1953. En 1977, le remplacement à la grandeur du réseau des étendues détaillées des couches nuageuses et des phénomènes obscurcissants par des états du ciel généraux fondés sur les étendues cumulatives a imposé l’analyse de Sea level responses to climatic variability and la fréquence des conditions principalement nuageuses change in Northern British Columbia plutôt que des étendues réelles. L’automatisation partielle ILUMIE S. ABEYSIRIGUNAWARDENA and IAN J. WALKER au moyen de systèmes d’observation météorologique D automatiques a entraîné la fin de l’observation du type de nuages et des nuages élevés de même qu’une incompatibilité de la couverture nuageuse avec les Abstract observations faites par des humains à huit pour cent des Sea level responses to climatic variability (CV) and change stations. (CC) signals at multiple temporal scales (inter-decadal to monthly) are statistically examined using long-term water Pour chaque observation horaire à 84 stations d’aéroports, level records from Prince Rupert (PR) on the north coast of de 1953 à 2003, chaque couche est classée, selon le type British Columbia. Analysis of observed sea level data from de nuage et la hauteur de base normalisée correspondante, PR, the longest available record in the region, indicates an dans trois niveaux : nuages bas, moyens et élevés. Les annual average mean sea level (MSL) trend of +1.4±0.6 -1 tendances dans l’occurrence des étendues cumulatives des mm yr for the period (1939-2003), as opposed to the -1 conditions principalement nuageuses à chacun de ces trois longer term trend of 1±0.4 mm yr (1909-2003). This niveaux sont calculées aux échelles annuelle, saisonnière, suggests a possible acceleration in MSL trends during the diurne et nocturne en même temps que les tendances latter half of the twentieth century. According to the results annuelles dans l’occurrence de nuages convectifs et of this study, the causes behind this acceleration can be stratiformes sélectionnés. D’après les anomalies annuelles attributed not only to the effects of global warming but also moyennées pour le pays et les provinces, aucune to cyclic climate variability patterns such as the strong discontinuité systématique importante dans l’ensemble du positive Pacific Decadal Oscillation (PDO) phase that has réseau n’a été trouvée; en moyenne, au cours d’une année been in existence since mid-1970s. The linear regression dans l’ensemble du réseau, on observe de faibles model based on highest sea levels (MAXSL) of each -1 tendances à la baisse pour les étendues cumulatives des calendar year showed a trend (3.4 mm yr ) exceeding twice conditions principalement nuageuses dans les niveaux bas that of MSL. Previous work shows that the influence of et moyens et des tendances à la hausse dans les niveaux vertical crustal motion on relative sea level is negligible at élevés. PR.

CMOS Bulletin SCMO Vol.36, No.3 -108- Relations between sea levels and known CV indices (e.g., sommes cumulées (CumSum) et l’analyse par époques the Multivariate ENSO Index (MEI), PDO, Northern superposées (SEA). Les résultats suggèrent que le forçage Oscillation Index (NOI), and Aleutian Low Pressure Index ENSO (comme le montrent les indices MEI et IOB) exercent (ALPI)) are explored to identify potential controls of CV une influence appréciable sur les fluctuations du niveau de phenomena (e.g., the El Niño Southern Oscillation (ENSO) la mer en hiver alors que l’ODP domine la variabilité du and PDO) on regional MSL and MAXSL. Linear and non- niveau de la mer en été. Les preuves observationnelles à linear statistical methods including: correlation analyses, PR montrent aussi que, durant la période 1939–2003, ces multiple regression, Cumulative Sum (CumSum) analysis, fluctuations cycliques du niveau de la mer de plus petite and Superposed Epoch Analysis (SEA) are used. Results échelle temporelle en réponse aux phénomènes de VC suggest that ENSO forcing (as shown by the MEI and NOI étaient nettement plus importante que la tendance à la indices) exerts significant influence on winter sea level hausse du niveau de la mer à plus long terme, par jusqu’à fluctuations, while the PDO dominates summer sea level un ordre de grandeur et avec des tendances plus de deux variability. The observational evidence at PR also shows fois plus grandes que celle du NMM. Des fluctuations du that, during the period 1939-2003, these cyclic shorter niveau de la mer aussi extrêmes liées aux événements de temporal scale sea level fluctuations in response to CV VC devraient constituer la priorité immédiate pour were significantly greater than the longer term sea-level rise l’élaboration de stratégies d’adaptation côtière, puisqu’elles trend, by as much as an order of magnitude and with trends se superposent aux tendances du NMM à long terme et over twice that of MSL. Such extreme sea level fluctuations aggravent les dangers qui résultent de la seule tendance à related to CV events should be the immediate priority for the la hausse du NMM à plus long terme. development of coastal adaptation strategies, as they are superimposed on long term MSL trends, resulting in greater hazard than longer term MSL rise trends alone. NOW AVAILABLE - DISPONIBLE MAINTENANT Résumé [traduit par la rédaction] Nous examinons, du point de vue statistique, les réponses du niveau de la mer aux signaux de variabilité climatique ATMOSPHERE-OCEAN 46-1 (VC) et de changement climatique (CC) à des échelles temporelles multiples (d’interdécennale à mensuelle) à l’aide des enregistrements de niveaux d’eau à long terme Anniversary Issue on Ozone de Prince Rupert (PR) sur la côte nord de la Colombie- Numéro anniversaire sur l’ozone Britannique. L’analyse des données d’observation du niveau de la mer de PR, la plus longue série d’enregistrements dans la région, indique une tendance # Ozone: From Discovery to Protection, annuelle moyenne du niveau moyen de la mer (NMM) de by C.T. MCELROY and P.F. FOGAL +1,4±0,6 mm an-1 pour la période (1939–2003) comparativement à la tendance à plus long terme de # Understanding Ozone Depletion: Measurements and -1 +1±0,4 mm an (1909–2003). Cela suggère une possible Models, by C.T. MCELROY and P.F.FOGAL accélération des tendances du NMM durant la deuxième moitié du XXe siècle. Selon les résultats de cette étude, on # Ozone Climatology, Trends, and Substances that peut attribuer les causes de cette accélération non Control Ozone, by V.E. FIOLETOV seulement aux effets du réchauffement planétaire mais aussi à des configurations de variabilité climatique # Stratospheric Ozone Chemistry, cycliques, comme la forte phase positive de l’oscillation by JOHN C. MCCONNELL and JIAN JUN JIN décennale du Pacifique (ODP) qui perdure depuis le milieu des années 1970. Le modèle de régression linéaire basé # Ozone in the Troposphere: Measurements, sur les plus hauts niveaux de la mer (MAXSL) de chaque Climatology, Budget, and Trends, by D.W. TARASICK and année civile a montré une tendance (3,4 mm an-1) de plus R. SLATER du double de celle du NMM. Des travaux précédents # Dynamics, Stratospheric Ozone, and Climate Change, montrent que l’effet des mouvements verticaux de la croûte by THEODORE G. SHEPHERD terrestre sur le niveau relatif de la mer sont négligeables à PR. # Connections between Stratospheric Ozone and Climate: Radiative Forcing, Climate Variability, and Nous explorons les relations entre les niveaux de la mer et Change les indices VC connus — p. ex. l’indice ENSO multivarié by N. MCFARLANE (MEI), l’ODP, l’indice d’oscillation boréale (IOB) et l’indice de dépression des Aléoutiennes (ALPI) — pour identifier # Surface Ultraviolet Radiation, l’influence potentielle des phénomènes de VC — p. ex. le El by J.B. KERR and V.E. FIOLETOV Niño—oscillation australe (ENSO) et l’ODP — sur le NMM et le MAXSL régionaux. Nous utilisons des méthodes # Effects of Ozone Depletion and UV-B on Humans and statistiques linéaires et non linéaires, notamment les the Environment, by KEITH R SOLOMON analyses de corrélation, la régression multiple, l’analyse des

-109- CMOS Bulletin SCMO Vol.36, No.3 IN MEMORIAM

Edward Lorenz, father of chaos theory and AM in mathematics from Harvard University in 1940, an SM butterfly effect, dies at 90 in meteorology from MIT in 1943 and an ScD in meteorology from MIT in 1948. It was while serving as a weather forecaster for the U.S. Army Air Corps in World War II that he April 16, 2008 decided to do graduate work in meteorology at MIT.

Edward Lorenz, an MIT meteorologist who tried to explain "As a boy I was always interested in doing things with why it is so hard to make good weather forecasts and wound numbers, and was also fascinated by changes in the up unleashing a scientific revolution called chaos theory, died weather," Lorenz wrote in an autobiographical sketch. April 16 of cancer at his home in Cambridge. He was 90. Lorenz was a member of the staff of what was then MIT's A professor at MIT, Department of Meteorology from 1948 to 1955, when he was Lorenz was the first to appointed to the faculty as an assistant professor. He was recognize what is now promoted to professor in 1962 and was head of the called chaotic behaviour department from 1977 to 1981. He became an emeritus in the mathematical professor in 1987. modelling of weather systems. In the early 1960s, Lorenz realized Predictability: Does the Flap of a Butterfly's Wings that small differences in a in Brazil Set Off a Tornado in Texas? dynamic system such as the atmosphere--or a Prof. Edward Lorenz model of the atmosphere- 1972 -could trigger vast and often unsuspected Lorenz, who was elected to the National Academy of results. Sciences in 1975, won numerous awards, honours and honorary degrees. In 1983, he and former MIT Professor These observations Henry M. Stommel were jointly awarded the $50,000 ultimately led him to Crafoord Prize by the Royal Swedish Academy of Sciences, formulate what became a prize established to recognize fields not eligible for Nobel known as the butterfly Prizes. Prof. Edward Lorenz in 1996 effect--a term that grew Photo courtesy of MIT News Office out of an academic paper In 1991, he was awarded the Kyoto Prize for basic sciences he presented in 1972 in the field of earth and planetary sciences. Lorenz was cited entitled: "Predictability: Does the Flap of a Butterfly's Wings by the Kyoto Prize committee for establishing "the theoretical in Brazil Set Off a Tornado in Texas?" basis of weather and climate predictability, as well as the basis for computer-aided atmospheric physics and Lorenz's early insights marked the beginning of a new field of meteorology." The committee added that Lorenz "made his study that impacted not just the field of mathematics but boldest scientific achievement in discovering 'deterministic virtually every branch of science--biological, physical and chaos', a principle which has profoundly influenced a wide social. In meteorology, it led to the conclusion that it may be range of basic sciences and brought about one of the most fundamentally impossible to predict weather beyond two or dramatic changes in mankind's view of nature since Sir Isaac three weeks with a reasonable degree of accuracy. Newton."

Some scientists have since asserted that the 20th century will During leaves of absence from MIT, he held research or be remembered for three scientific revolutions--relativity, teaching positions at the Lowell Observatory in Flagstaff, quantum mechanics and chaos. Arizona; the Department of Meteorology at the University of California at Los Angeles; Det Norske Meteorologiske Institutt "By showing that certain deterministic systems have formal in Oslo, Norway; and the National Center for Atmospheric predictability limits, Ed put the last nail in the coffin of the Research in Boulder, Colorado. Cartesian universe and fomented what some have called the third scientific revolution of the 20th century, following on the An avid hiker and cross-country skier, Lorenz was active up heels of relativity and quantum physics," said Kerry Emanuel until about two weeks before his death, his family said. professor of atmospheric science at MIT. "He was also a Lorenz is survived by three children, Nancy, Edward and perfect gentleman, and through his intelligence, integrity and Cheryl, and four grandchildren. His wife, Jane, died in 2001. humility set a very high standard for his and succeeding generations." Source: MIT website visited April 18, 2008. Reprinted here Born in 1917 in West Hartford, Connecticut, Lorenz received with the written authorization of the MIT News Office. an AB in mathematics from Dartmouth College in 1938, an

CMOS Bulletin SCMO Vol.36, No.3 -110- SHORT NEWS / NOUVELLES BRÈVES

7th International Workshop on Unstructured Grid Abstracts should have the form: Modelling of Coastal, Shelf and Ocean Flows 1) Title 2) Authors/affiliations/address(email & postal) 3) Abstract (250 word max, no figures).

If using MSWord, font should be 12pt, Times New Roman.

In addition to the usual presentations, discussion sessions and posters, present plans call for a half day session summarizing the state of the art in unstructured modelling that will be open to the wider Halifax scientific community.

We anticipate a small workshop fee that will cover coffee breaks, lunches served on site and the workshop dinner. Receipts will be issued. Weekend expeditions and tours are being investigated. The workshop receives support from Department of Fisheries and Oceans (DFO) and Canadian Meteorological and Oceanographic Society (CMOS). The Seventh International Workshop on Unstructured Grid Modelling of Coastal, Shelf and Ocean Flows will Previous workshop locations included: be held at the Bedford Institute of Oceanography in Halifax, Nova Scotia, September 17-19 2008. # 2002: Liege, Belgium. # 2003: Delft, Netherlands. The workshop website is: # 2004: Toulouse, France # 2005: Bremerhaven, Germany. # 2006: Miami, USA. http://wessex.eas.ualberta.ca/~myers/unstructgrid # 2007: London, England. 2008 Previous Workshop Publications: All announcements for the workshop will be made via the mailing list: 1) Ocean Modelling Special Issue: The second international workshop on unstructured mesh numerical modelling of coastal, [email protected] shelf and ocean flows, Delft, The Netherlands, September 23- 25, 2003. and to people who have indicated that they wish to 2) Ocean Modelling Special Issue: The third international receive separate notices. workshop on unstructured mesh numerical modelling of coastal, shelf and ocean flows, Toulouse, France, September 20-22, For more information, see: 2004.

http://wessex.eas.ualberta.ca/~myers/FE/fe- 3) Ocean Modelling Special Issue: The sixth international list.html. workshop on unstructured mesh numerical modelling of coastal, shelf and ocean flows, London, England, September 19-21, If anybody would like to receive further notices on the 2007. In prep. workshop but does not want to subscribe, please contact David Greenberg at US - Canada Agreement to Work Together on Weather and Climate Research [email protected]. New Orleans, Louisiana, United States, January 22, 2008 - The Abstract Submission National Oceanic and Atmospheric Administration (NOAA) and Abstracts should be emailed in plain text or MSWord to Environment Canada signed a cross-border agreement to enhance weather and climate monitoring and research. [email protected] "With this agreement, NOAA and Environment Canada can broaden our collaboration to enhance health, safety and by June 15, 2008. economic prosperity for our countries and the world," said retired

-111- CMOS Bulletin SCMO Vol.36, No.3 Navy Vice Adm. Conrad Lautenbacher, Ph.D., under protect our ocean and coastal resources," said Robert secretary of commerce for oceans and atmosphere and Gagosian, President and CEO of the Consortium for Ocean NOAA administrator. "Although our organizations have Leadership. The Integrated Ocean Observing System was engaged in a successful relationship for decades, we identified as a central ocean priority by the U.S. Commission on now have a framework that allows us to explore new Ocean Policy (2003), the Pew Oceans Commission (2004), the opportunities." Joint Ocean Commission Initiative (2006) and the Ocean Research Priorities Plan and Implementation Strategy (2007). The Memorandum of Understanding establishes a "Congress took a significant step toward making these formal link between the two organizations, as well as a recommendations a reality, which will fundamentally change standing committee with representatives from both how we study the ocean and transfer information into products countries that will meet regularly to collaborate on and services for the public," continued Gagosian. cross-border projects. The IOOS bill (H.R.2342) passed the House by voice vote. The "International partnerships like this between Senate has yet to consider its version of the IOOS bill (S.950), meteorological services is key to advancing the quality which was approved by the Senate Commerce, Science and of our weather and environmental forecasting services Transportation Committee last year. protecting citizens' health and property," said Mark Warawa, Parliamentary Secretary to Canada's Environment Minister John Baird. "This agreement is Proceedings of IUGG XXIV General Assembly another important milestone in our efforts to understand Perugia, Italy, 2007 and combat climate change." All abstracts presented at the XXIV IUGG General Assembly, held in Perugia from July 2nd to the 13th, 2007 are now available on line, including those of IAPSO. The abstracts as pdf files can House Passes Critical Ocean Observing System be viewed and/or downloaded from the www.iugg2007perugia.it Bill web site at the following link: http://www.iugg2007perugia.it/webbook/ (Washington, D.C.) - On April 1, the U.S. House of Representatives passed the National Integrated

Coastal and Ocean Observing Act of 2007. The Act Second Symposium on The Ocean in a High-CO2 World would create an Integrated Ocean Observing System Registration and abstract submissions are now open for the

(IOOS) that will monitor and forecast ocean conditions, Second Symposium on “The Ocean in a High-CO2 World” to be including the physical, biological and chemical held in Monaco on October 6-9, 2008. The symposium web site components of coastal waters. The Act would also can be found at http://www.highco2world-ii.org. increase understanding of complex deep ocean and coastal environments and promote the dissemination of information to local policymakers and the public. CMOS Accredited Consultants An integrated ocean observing system will protect Experts-Conseils accrédités de la SCMO public health through the identification of marine toxins and pollutants in coastal areas as well as improving Gamal Eldin Omer Elhag Idris, C.Chem., MCIC warnings of tsunamis, hurricanes, El Niño events and other natural hazards. The system will also enhance Chemical Oceanography, homeland security, support maritime operations and Pollution Control and Water Technology collect important information needed to address global warming, improve ocean health and provide for the 211-100 High Park Avenue protection, sustainable use and enjoyment of ocean Toronto, Ontario M6P 2S2 Canada resources. Tel: 416-604-9165 (Home) Email; [email protected] "The ocean plays an integral role in our economy, environment and quality of life. Yet our understanding of the ocean and coastal environment is very limited," Douw G. Steyn said Representative Thomas Allen (D-ME) who introduced the bill. "We need the equivalent of the national weather service for the ocean and IOOS Air Pollution Meteorology provides us with that predictive capability for the Boundary Layer & Meso-Scale Meteorology oceans," continued Representative Allen. 4064 West 19th Avenue Vancouver, British Columbia, "The ocean science community has been working for more than a decade on developing observing systems, V6S 1E3 Canada which will help us better understand, manage and Tel: 604-822-6407; Home: 604-222-1266

CMOS Bulletin SCMO Vol.36, No.3 -112-

OUR PEOPLE, YO U R DATA notre personnel, vos données

www.campbellsci.ca